For most organizations, including federal contractors and subcontractors, workforce data is the foundation of legal compliance. The Office of Federal Contract Compliance Programs, Equal Employment Opportunity Commission, and a host of related regulations require that contractors maintain detailed, accurate, and organized records on their workforce. Done well, data collection shields your organization during audits and drives more equitable employment decisions. Done poorly, it exposes you to significant legal liability.
Demographic data is required for EEO-1 reporting and certain state compliance reporting (Minnesota affirmative action plans, California pay reporting, etc.). Despite being familiar to most employers, it is also one of the most commonly mishandled.
Federal and state regulations require that race and sex information be collected through employee and applicant self-identification wherever possible. Employers are discouraged from making visual determinations or "best guesses" about an individual's race or sex. If an employee declines to self-identify, the employer is permitted to make a visual observation, but this must be clearly documented as a secondary method, never the first. Visual identification should also be completed by the employee’s direct supervisor when possible.
The EEO-1 report currently allows for only binary sex data (male/female), with the recommendation to note non-binary individuals manually for each establishment. However, best practice increasingly means providing a third, non-binary option for internal self-identification forms. As federal and state reporting categories evolve, having the more granular data positions organizations to adapt.
Perhaps no category of data is more consequential (or more problematic in practice) than applicant flow data. Analysis of applicant flow logs can indicate whether the organization’s selection processes are operating equitably. While federal contractors are no longer required to complete applicant selection rate analyses under Executive Order (EO) 11246, new EOs released in 2025 and 2026 require that contractors certify they are not violating federal anti-discrimination laws or engaging in “racially discriminatory DEI activities”. Completing robust workforce analytics, including of your applicant data and selection rates, is the best way to check that your selection policies and procedures do not have unintended consequences.
Thorough and accurate applicant data is critical to being able to complete meaningful data analyses. Organizations should capture, at a minimum, the following information for every applicant:
Organizations are highly encouraged to use standardized disposition codes across all requisitions. Allowing free-text "rejection reasons" can leave organizations with data that is impossible to analyze systematically, and that can expose legally problematic language if reviewed during a lawsuit or government review.
Employment decisions made after hire are just as legally significant as the initial selection. Discrimination in promotion, transfer, demotion, and termination is equally actionable. This means data systems must capture the full arc of an employee's tenure.
Data points that should be captured for personnel movements includes:
Promotions — effective date, prior title/grade, new title/grade, selecting official
Lateral transfers — requested vs. employer-initiated, effective date
Demotions — reason, effective date, whether voluntary or involuntary
Terminations — voluntary or involuntary, reason, last day worked
Pay transparency has become the primary focus of many state regulations. Maintaining detailed, accurate compensation data is no longer optional, it is a core compliance function.
What to maintain in your compensation record:
Base salary or hourly rate, with effective date of each change
Reason for each pay change (hire, merit increase, promotion, market adjustment, equity adjustment)
Starting salary at time of hire and compa-ratio at hire
Bonus and variable compensation
Pay grade, band, or level
Total compensation
Pay increase history
The single most important thing a company can do to build defensible compensation records is to require documented rationale for every pay action, especially at hire and promotion. This documentation does not need to be lengthy. A structured compensation approval form with a few mandatory fields is far more defensible than a system that simply records a number with no explanatory context. Organizations are highly encouraged to conduct annual proactive compensation equity analyses, where pay adjustments can be made if necessary.
Performance ratings are important data points because they are often relied on for merit increases, promotion decisions, reductions in force, and terminations. If performance ratings themselves are biased or unreliable, then every downstream decision that relies on them may also be tainted.
Organizations should periodically run distribution analyses on performance ratings by race, sex, and other relevant demographic characteristics. Consistently lower average ratings for one group, absent documented, job-related explanatory factors, can become a legal liability.
Under Section 503 of the Rehabilitation Act of 1973 and the Vietnam Era Veterans Readjustment Assistance Act (VEVRAA) , federal contractors must invite applicants and employees to voluntarily self-identify as individuals with disabilities and as protected veterans. These invitations must occur at three points in the employment lifecycle: pre-offer, post-offer, and at least every five years for disability status.
Contractors must use the Office of Management and Budget-approved self-identification form for disability. Disability self-identification responses must be stored separately from personnel files. Along with completing the analyses required under Section 503 and VEVRAA, contractors should also track overall response rates, as low response rates could indicate another issue.
Data collection is not a passive exercise in paperwork. It is the evidentiary record of how your organization treats its people. Companies that invest in complete, consistent, and well-governed workforce data are better equipped to defend their employment decisions, identify and correct internal inequities before they compound, and maintain the government contracts on which their business depends. Treat data collection not as a compliance cost, but as the infrastructure for fair and defensible employment practices.