Understanding the topic
A program audit is broader than reviewing a few reports. It examines why screening occurs, which workers and locations are covered, what services are ordered, how candidates are notified, who can access information, how decisions are documented, and whether technology and vendor configurations match policy.
The strongest audits use evidence. They compare policy, templates, system settings, user lists, sample orders, access logs, dispute files, adverse-action records, retention practices, invoices, and interviews with the people performing the work.
Audit domains to include
The scope should follow the complete lifecycle from role design through report disposal.
- Governance, policy ownership, legal review, covered populations, locations, and program purpose.
- Role matrix, packages, search scope, optional components, client requirements, and exceptions.
- Disclosure, authorization, candidate data collection, state notices, timing, and certifications.
- Report review, identity matching, job relevance, individualized assessment, adverse action, and disputes.
- Account access, integrations, privacy, security, retention, disposal, vendor oversight, billing, and metrics.
A practical audit method
An audit should identify both isolated errors and systemic causes.
- 1
Define scope and criteria
Identify business units, roles, locations, worker types, systems, vendors, and the policies or laws used as audit criteria.
- 2
Collect configuration and records
Obtain forms, packages, users, integration mappings, sample files, reports, notices, logs, invoices, and training materials.
- 3
Test representative scenarios
Trace candidates through clear, delayed, disputed, record-found, multi-state, driving, credit, and adverse-action workflows.
- 4
Classify findings
Separate urgent legal or security issues, recurring control gaps, one-off errors, and improvement opportunities.
- 5
Remediate and verify
Assign owners and deadlines, change policy and systems, retrain users, and retest the corrected workflow.
Audit mistakes to avoid
A checklist alone can miss actual behavior and technical configuration.
- Reviewing only written policy without observing recruiters and managers.
- Sampling only completed clear reports and excluding disputes, exceptions, or adverse decisions.
- Allowing the vendor to define all audit criteria without employer legal and operational input.
- Reporting findings without an owner, deadline, evidence of correction, or retest.
- Treating every location and worker type as though the same forms and rules apply.
Evidence and metrics to review
Data can reveal where the program differs from policy, but metrics should not be used as hiring quotas.
- Package use by role and location, including manual additions and exceptions.
- Candidate completion, validation errors, expired invitations, source delays, and support contacts.
- Report access, exports, former-user accounts, shared credentials, and unusual order patterns.
- Disputes, corrected reports, pre-adverse cases, final decisions, and process timing.
- Third-party fees, duplicate orders, cancelled services, billing variance, and integration failures.
Putting the information into a screening program
An audit should not require the employer to retain screening data longer than its reviewed schedule or copy sensitive reports into an insecure audit folder. Use controlled access, minimal necessary data, and a documented destruction plan.
Findings should be prioritized by candidate impact, legal and security risk, frequency, and operational effect. The final report should distinguish confirmed issues from items requiring legal interpretation.
Important legal and jurisdictional note
PRIMARY SOURCES
Sources and further reading
- Background Checks: What Employers Need to KnowFederal Trade Commission
- Background Checks: What Employers Need to KnowU.S. Equal Employment Opportunity Commission
RELATED SERVICES
