Our series on screening filters has covered degree requirements, employment gap rules, tenure rules, job title matching, and location and radius limits. Each one is a rule that decides who a reviewer sees. This post gathers them into a single audit. For the research behind why these filters matter, start with the Hidden Workers overview.

The Legal Reference Points

Federal selection guidelines. The Uniform Guidelines on Employee Selection Procedures, at 29 CFR 1607.4(D), describe the four-fifths guideline: "A selection rate for any race, sex, or ethnic group which is less than four-fifths ( 4/5 ) (or eighty percent) of the rate for the group with the highest rate will generally be regarded by the Federal enforcement agencies as evidence of adverse impact." The same paragraph adds: "Smaller differences in selection rate may nevertheless constitute adverse impact, where they are significant in both statistical and practical terms." It is a screening guideline, not a safe harbor.

Federal enforcement priorities. Foley & Lardner reported on June 15, 2026, that "On June 4, 2026, the Equal Employment Opportunity Commission (EEOC) issued a revised National Enforcement Plan" and that "The agency further stated that it will prioritize claims of disparate treatment over disparate impact." A plan describes where the agency intends to focus. It does not amend the statute, and it does not bind private plaintiffs or state agencies. How much weight to give it is a question for employment counsel.

California. The Civil Rights Council's automated-decision system regulations took effect on October 1, 2025. The Civil Rights Department's announcement of the rules describes the goal of having employers "maintain employment records, including automated-decision data, for a minimum of four years." The regulations define an automated-decision system to include screening resumes for particular terms or patterns, and they make evidence of anti-bias testing relevant to a claim or defense. Our post on California automated-decision compliance covers the details.

New York City. Local Law 144 is a reference point even for employers outside the city. The city's page says a tool must be "subject to a bias audit within one year of the use of the tool," with information about the audit publicly available, and that enforcement began on July 5, 2023.

Litigation. Mobley v. Workday, Inc., filed in the Northern District of California on February 21, 2023, is a putative class action over a vendor's applicant screening tools. The docket remained open as of October 2026, with an amended complaint filed on July 10, 2026. A pending case is not a ruling, but it shows that software vendors can be named in disputes over screening tools.

The Audit, Step by Step

Step 1. Inventory every automated step. List knockout questions, degree and certification rules, gap and tenure rules, title matching, location limits, keyword scores, and any ranking or auto-reject threshold. Ask the vendor for a configuration export instead of relying on memory. Our post on knockout questions explains why these are easy to miss.

Step 2. Write the job-related reason for each rule. One or two sentences stating what the rule measures and why the role needs it. A rule with no stated reason is the first candidate for removal or replacement.

Step 3. Collect step-by-step outcomes. For each requisition, record how many applicants entered and how many passed each automated step. Where you collect voluntary self-identification data, ask counsel what you may collect and how it may be used in this analysis.

Step 4. Calculate selection rates and impact ratios. Divide the number who passed by the number who entered, for each group. Divide each group's rate by the highest group's rate. A hypothetical example: if 120 of 200 applicants in one group pass a gap rule (60%) and 42 of 100 in another group pass (42%), the ratio is 0.70, which is below the four-fifths line and calls for a closer look.

Step 5. Look beyond the ratio. The guideline itself warns that small numbers can mislead in both directions. Review the sample size and statistical significance, and ask whether the difference is meaningful in practice.

Step 6. Review a sample of rejections. Pull resumes that the system rejected and have a reviewer judge them against the posting. This test catches filters that pass the numbers but still reject capable people.

Step 7. Fix or document each finding. Remove the rule, change it to a flag, or record why it stays and who approved it. The California regulations make the quality and recency of testing and the response to results relevant, so keep a note of what you did after each review.

Step 8. Retain the records. California's four-year period covers automated-decision data, and it is a sensible baseline for employers elsewhere to raise with counsel. Keep configurations, outcome counts, review notes, and approvals together.

Step 9. Repeat on a schedule. Run the audit again after any change to the ATS, the postings, or the vendor's scoring, and at least once a year.

Step 10. Take it to counsel. Share the inventory and the results with employment counsel before relying on them, and ask which state and local rules apply.

Questions to Put to Your Vendor

How TrueScan HR Helps

TrueScan HR scores a resume against the job description and shows the matched and missing skills, which gives an audit a concrete record of why each candidate moved forward or did not. A narrative assessment explains each score in plain language, so the reasoning can be reviewed and questioned. Veteran Mode translates military codes, duties, and clearances into civilian skills, and the reviewer makes the decision at every step.

This post is general information, not legal advice. TrueScan HR is not a law firm. Confirm how these rules apply to your organization with employment counsel.

Sources: