U.S. Public Safety HR: 7 Steps to Build an Adjudication Matrix

An adjudication matrix is a standardized grid that maps criminal record findings to hiring decisions by offense type, lookback period, and role risk tier. The governing rule is simple but non-negotiable: it functions as a guide, never as an automatic disqualifier. Under EEOC and FCRA rules, every gray-area result still requires an individualized assessment, documented and defensible, before OMNI Intel or any employer finalizes an adverse decision.

TL;DR:

• Only green cells allow for automatic clearance; yellow and red results require documented human review and detailed individual assessments.
• Lookback periods should be tailored to each offense category and role risk tier, with clear written reasoning to defend decisions if challenged.
• Legal compliance demands applying the EEOC’s Green factors and FCRA pre-adverse action notices, with special attention to jurisdiction-specific timing laws.
• Regularly pilot and review the matrix against historical cases to calibrate offense categories and lookback windows, involving legal counsel early in development.
• Automated filtering should exclude irrelevant or sealed records, but cases in the yellow and red zones must always undergo human adjudication to avoid legal and compliance issues.

Table of Contents

• What Does an Adjudication Matrix Contain?
• What Legal Rules Govern an Adjudication Matrix?
• How Do You Build an Adjudication Matrix Step by Step?
• How Do You Run an Adjudication Matrix Day to Day?
• What Does a Sample Adjudication Matrix Look Like?
• How OMNI Intel Supports Adjudication for Public-Safety Hiring
• Perspective: Common Pitfalls and the Fixes to Prioritize
• OMNI Intel: a Vendor Built for Public-Safety Adjudication
• Sources
• FAQ

What Does an Adjudication Matrix Contain?

A matrix built for adjudication matrix hiring decisions works because it forces three separate variables onto one grid instead of letting a reviewer eyeball a rap sheet and guess. Skip one axis and the whole document becomes a liability rather than a shield.

Offense categories group findings by type: violent crime, theft or fraud, drug offenses, traffic violations, and so on. Broad “any felony” buckets invite legal trouble because they ignore job relevance, one of the core factors regulators expect employers to weigh.

Lookback periods set how far back a given offense category counts. A dispatch center might look back seven years for theft convictions but treat a decades-old misdemeanor as irrelevant. Lookback windows should track the role’s actual exposure to risk, not an arbitrary agency-wide number pulled from a template.

Role risk tiers rank positions by responsibility and public trust. A patrol officer candidate and a records clerk candidate should never run through identical rules, because their access to weapons, vulnerable people, and sensitive data differs enormously.

Most matrices then resolve to a simple three-color decision system:

• Green (eligible): the offense category and timeframe fall well outside any relevance to the role; the matrix clears the candidate to proceed.
• Yellow (individualized review required): the finding sits close enough to the role or timeframe that a human adjudicator must apply the Green factors before any decision is made.
• Red (likely ineligible, pending individualized assessment): the offense is serious, recent, and directly job-related, but even here the candidate typically gets a chance to respond before a final decision is locked in.

Only green cells should route to fully automated clearance. Every yellow and red cell needs a named human reviewer, a timestamp, and written reasoning before the case closes. That routing logic, more than any single offense list, is what separates a defensible matrix from a lawsuit waiting to happen.

What Legal Rules Govern an Adjudication Matrix?

Three legal frameworks constrain how a matrix can operate, and ignoring any one of them turns a well-designed grid into evidence against the employer that built it.

The EEOC’s enforcement guidance requires that criminal record screening policies be job-related and consistent with business necessity. To meet that standard, the agency recommends applying what practitioners call the Green factors:

1. Nature and gravity of the offense. A theft conviction weighs differently for a cash-handling role than for a warehouse position with no financial access.
2. Time elapsed since the offense or completion of the sentence. A conviction from many years ago carries less weight than a recent one, absent a pattern of repeat conduct.
3. Nature of the job held or sought. A DUI from three years back matters more for a role requiring a commercial driver’s license than for a desk-based records position.

FCRA compliance runs on a separate but parallel track. The FTC’s guidance requires employers to send a pre-adverse action notice, including a copy of the consumer report and a summary of rights, before taking any adverse action based on a background check. Fisher Phillips frames this as a four-step plan: send the pre-adverse notice, treat the matrix result as a starting point rather than a verdict, give the candidate a real opportunity to respond or dispute inaccuracies, then issue the final adverse-action notice, complete with the screening company’s contact information, only if the decision stands.

A defensible yellow-tier record documents rehabilitation, time elapsed, and job relevance, with a logged adjudicator decision. That documentation is what a court or an EEOC investigator will ask to see first if a rejected candidate challenges the outcome, according to LegalClarity’s breakdown of how these workflows hold up under scrutiny.

State and local “ban the box” laws add a third layer that many multi-state agencies underestimate. These laws restrict when in the hiring process an employer can even ask about criminal history, and some cities require additional individualized-assessment steps beyond federal law. OMNI Intel’s ban the box guide breaks down the state-by-state timing rules, which matters because a matrix that triggers a background check before a conditional offer in a jurisdiction that forbids it is a compliance violation regardless of how fair the matrix itself is.

Before any matrix goes live, run it through a short compliance checklist: validate offense categories against actual job duties, get legal counsel to sign off on lookback periods, confirm the matrix encodes every jurisdiction’s ban-the-box timing correctly, and set a record retention schedule that meets both FCRA and state requirements.

How Do You Build an Adjudication Matrix Step by Step?

Building a matrix from scratch takes longer than downloading a template, but a template built for someone else’s risk tiers will misfire on your roles. Here’s the sequence that produces something legally sound and operationally usable.

1. Map every role to a risk tier. Start by listing essential functions and exposure points. Does the role involve firearms, minors, unsupervised access to vulnerable people, or financial systems? A 911 dispatcher and a fleet mechanic at the same agency belong in different tiers even if both report to the same department.
2. Select offense categories per tier. Pull categories from actual incident data and role requirements, not a generic industry list. A private security firm screening armed guard candidates needs a more granular violent-offense category than a records department screening clerical staff.
3. Set lookback windows per category and tier. Document the reasoning behind each window in writing. “Seven years for theft offenses in cash-handling roles” is defensible; “we’ve always used seven years” is not.
4. Define cell outcomes and escalation rules. Specify exactly which offense-and-timeframe combinations trigger automatic clearance, which require enhanced individualized review, and who owns that escalation.
5. Build the individualized-assessment worksheet. This is the single artifact regulators will ask for during any dispute, so it needs defined fields, not a blank text box.
6. Pilot the matrix on anonymized historical files. Run last year’s applicant pool through the draft matrix before touching a live candidate, and check whether the outcomes match what your legal team expects.
7. Schedule an annual review. Fair-chance laws change frequently at the state and municipal level; a matrix built in 2024 may already be out of step with a new 2026 local ordinance.

The individualized-assessment worksheet should capture, at minimum:

• Candidate’s written explanation of the offense and circumstances
• Evidence of rehabilitation (certificates, employer references, time in stable employment)
• Specific connection, or lack of one, between the offense and the job’s essential functions
• Reviewer name, date, and final reasoning in the candidate’s own case file

Pro Tip: Pilot the matrix against a sample of last year’s actual adjudications before rolling it out. If the new matrix had flipped outcomes for candidates you already hired without incident, that’s a signal your lookback windows or offense categories are miscalibrated, not a signal to ignore.

Loop legal counsel in at the tier-mapping stage, not after the matrix is built. A compliance review that only happens at the end catches errors after they’ve already shaped every downstream decision. OMNI Intel’s key compliance steps guide walks through where legal review should sit in that sequence for public-safety agencies specifically.

How Do You Run an Adjudication Matrix Day to Day?

A matrix on paper means nothing without clear ownership of who touches a case at each stage. Most defensible workflows split the process into four roles: a screener who pulls and verifies raw records, an adjudicator who applies the matrix and runs individualized assessments on yellow and red cells, a legal escalation contact for disputed or unusual cases, and a final approver who signs off before any adverse-action notice goes out.

The end-to-end workflow typically looks like this: a background check returns, automated logic sorts the finding into green, yellow, or red, green cases clear without human intervention, and yellow or red cases route to a named adjudicator who completes the individualized-assessment worksheet before any notice is drafted.

Automation earns its place here by filtering out irrelevant findings, sealed records, or expunged charges before a human ever sees the file, which reduces both workload and the risk of a reviewer being influenced by information that legally shouldn’t factor into the decision. Checkr’s guidance on adjudication makes the trade-off explicit: filtering irrelevant items reduces bias, but stripping out human review for borderline cases raises disparate-impact risk instead of lowering it. The fix is not less automation across the board. It’s automation scoped tightly to the green tier while yellow and red cases keep a human decision-maker every time.

Audit logging closes the loop. Every adjudication decision needs a timestamp, the reviewer’s identity, the factors considered, and the final outcome stored in a retrievable format. Build these habits into the routine:

• Log every yellow and red decision with reviewer name, date, and written reasoning, not just a pass or fail flag.
• Retain pre-adverse and adverse-action notices with proof of delivery for the period your state’s record-keeping rules require.
• Run an internal audit at least twice a year, checking a sample of closed cases against the written matrix rules.
• Flag any case where the final decision deviated from the matrix’s default outcome and require a documented justification.

OMNI Intel’s FCRA compliance guide for public-safety HR covers how to build these notice and retention steps directly into a hiring platform instead of tracking them in a separate spreadsheet.

What Does a Sample Adjudication Matrix Look Like?

A role-agnostic template lists offense categories across the top and lookback windows down the side, with each cell holding a decision code rather than a paragraph of policy language. SmartSMSSolutions’ compliance template uses exactly this layout, and it adapts easily once you swap in your own tiers.

Every yellow and red cell still routes to the individualized-assessment worksheet before any final decision.

The individualized-assessment checklist, keyed to the Green factors, should confirm:

• Nature and gravity of the offense documented against the specific role’s duties
• Time elapsed since the offense or sentence completion recorded precisely
• Connection between the offense and the essential functions of the job stated in writing
• Candidate’s response, evidence of rehabilitation, or dispute logged in the file

For every completed case, store the pre-adverse notice with delivery timestamp, the candidate’s response if one was submitted, the adjudicator’s written reasoning, and the final adverse-action notice if issued.

How OMNI Intel Supports Adjudication for Public-Safety Hiring

OMNI Intel builds tailored screening packages for law enforcement, fire and EMS, dispatch, and private security agencies, integrated directly with the hiring platforms these agencies already use. That integration matters because a matrix that lives in a separate spreadsheet from the applicant tracking system tends to fall out of sync the first time a policy changes.

AI-assisted filtering handles the volume work: sorting green-tier clearances and flagging yellow-tier cases for a human adjudicator rather than trying to automate a judgment call the EEOC expects a person to make. There is also adverse-action guidance and screening templates built specifically around public-safety investigation principles, so agencies are not adapting a generic corporate template to fit badge-and-gun hiring standards.

Perspective: Common Pitfalls and the Fixes to Prioritize

The mistake I see most often isn’t a badly designed matrix. It’s treating a well-designed one as the final word. Agencies build a solid grid, then let a yellow-cell result auto-deny a candidate because nobody assigned a human to actually do the individualized assessment. Inconsistent documentation is the second failure: two adjudicators handling similar cases differently, with no written reasoning to explain why.

Three fixes matter more than any others. Get legal counsel to review the matrix before launch, not after a complaint. Pilot it against anonymized historical cases to catch miscalibrated lookback windows early. Train every adjudicator on the Green factors specifically, since a reviewer who doesn’t understand why time elapsed matters will apply it inconsistently across candidates.

— Matt

OMNI Intel: a Vendor Built for Public-Safety Adjudication

Building a compliant matrix in a spreadsheet is possible. Running it consistently across every dispatch, patrol, and fire department opening your agency posts is where most homegrown systems break down. OMNI Intel gives public-safety HR teams a screening platform that integrates directly with existing hiring workflows, applies custom-built adjudication matrices tuned to your specific risk tiers, and pairs automated routing with human review on every yellow and red case, so nothing clears without the individualized assessment the EEOC expects.

The result is faster hiring cycles without sacrificing the documentation trail that protects your agency if a decision is ever challenged. Agencies get specialized screening built around public-safety roles specifically, not a repurposed corporate template. If your current process still relies on manual spreadsheets and inconsistent adjudicator notes, start by reviewing OMNI Intel’s pre-employment screening services to see how a tailored package would map onto your existing risk tiers.

Sources

The EEOC’s enforcement guidance sets the individualized-assessment standard every matrix must meet. The FTC’s FCRA guidance governs the pre-adverse and adverse-action notice requirements. For jurisdiction-specific timing rules, OMNI Intel’s ban the box guide and pre-employment background check guide offer practical templates built for public-safety hiring teams navigating multi-state operations.

This article is general information, not a substitute for advice from a qualified lawyer. Consult a qualified legal professional about your own circumstances before acting on anything here.

• Fair Credit Reporting Act (FCRA) | Federal Trade Commission
• Avoiding Hidden Hiring Landmines: 4-Step FCRA Compliance Plan | Fisher Phillips
• Background check adjudication matrix: how it works | LegalClarity

FAQ

What Is an Adjudication Matrix in a Background Check?

It’s a standardized grid that maps criminal record findings, by offense category, lookback period, and role risk tier, to a preliminary hiring decision. It guides the reviewer but never replaces the individualized assessment EEOC guidance requires for gray-area results.

What Disqualifies You From a Level 3 Background Check?

Disqualification criteria vary by state and by the specific role’s licensing requirements, since “Level 3” designations differ across jurisdictions. Generally, offenses directly related to the position’s duties, especially those involving violence, fraud, or crimes against vulnerable populations, carry the most weight in a properly built matrix.

What Counts as a Red Flag on a Background Check for Employment?

A red flag typically means a finding falls into a high-risk offense category within a recent lookback window for that specific role, triggering the matrix’s red or yellow tier. It should still prompt an individualized assessment rather than an automatic rejection, per FCRA and EEOC requirements.

How Does an Automated Background Adjudication System Work?

Automated systems typically clear green-tier cases outright based on the matrix rules, then route yellow and red-tier findings to a human adjudicator for individualized review. Checkr’s guidance notes that removing human review from borderline cases increases legal risk rather than reducing it, which is why platforms like OMNI Intel preserve human sign-off on every non-green result.

Do State Fair-Chance Laws Change How a Matrix Should Work?

Yes. Ban-the-box and fair-chance laws control when in the hiring process a background check inquiry can happen, and some jurisdictions add extra individualized-assessment steps beyond federal requirements. A multi-state matrix needs jurisdiction-specific timing rules built in, not a single national default.

Recommended

• Step by Step Screening Process for Public Safety Hiring
• What Is Adverse Action In Hiring? A Guide For Public Safety HR
• Public Safety Compliance Guide For HR Managers
• 7 Best Hiring Process Practices for Public Safety Agencies