Candidate research is ethical when it is job-related, applied the same way to every candidate, disclosed to candidates and documented. For a large corporate talent function, that means a written policy that sets what recruiters may look at, when they may look, who does it and how the results are recorded. Individual judgment alone does not scale across a team, and it is hard to defend if a hiring decision is challenged. Talent leaders in peer networks such as TLIX compare how their policies work in practice, which is where this topic is best tested.
This guide covers sourcing ethics (how candidates are found and approached) and candidate research (what is checked about them once they are in the process). It is written for heads of talent acquisition, Chief People Officers and executive recruiting leaders who own the policy and answer for it.
TL;DR
- Ethical research rests on four things: job relevance, consistency across candidates, candidate notice and a written record of what was found and how it was used.
- Keep searches to professional, public sources and run them late in the process, after an initial assessment. Separate the person who researches from the person who decides.
- Source from consented and opt-in channels. Reading a public profile and sending one personal note is normal; bulk copying profiles into a private database is not.
- US rules differ by state and city. Federal background-check guidance from the EEOC has been rescinded, while state and local criminal-history laws keep expanding.
- A defensible policy covers scope, notice, adverse-action steps, vendor oversight, training and retention.
Table of Contents
- What candidate research ethics covers
- Why it matters beyond compliance
- What to research, and what to leave out
- Sourcing ethics: consent is the dividing line
- The practical risks of informal searching
- Keeping research consistent and fair
- The US rules, and what has changed
- Building the policy
- Where training and peer review help
- Our perspective
- FAQ
- Sources
What candidate research ethics covers
Two activities sit under this heading.
Sourcing is finding and approaching people who have not applied. The ethical questions are about consent and honesty: where the data came from, whether the person expected to be contacted, and whether the pitch is accurate.
Candidate research is looking into an applicant or finalist: searching their online presence, running background checks, checking references, including informal ones.
Both carry the same risk. The recruiter gets the benefit, and the candidate carries the cost of privacy loss, bias or a decision based on information they never saw.
One ethics paper on recruiting and selection makes a point that is easy to miss: ethics starts with the position requirements. Unnecessarily stringent requirements that exclude qualified people are as defective as vague ones, and the same paper describes misrepresenting the scope, difficulty or pay of a role during sourcing as unfair. Research ethics therefore begins before the first search, with an honest and accurate brief.

Why it matters beyond compliance
Cybervetting, meaning screening candidates through internet and social media content, is common. A 2021 summary of research reports that roughly 7 in 10 US workplaces use it. The same source set cites a 2015 SHRM survey in which 43% of HR professionals cybervet, and a 2018 CareerBuilder study in which 70% of hiring managers and HR professionals combined do.

The evidence on its value is weaker than its use. A 2022 academic review reports that assessments of online content have little predictive value for job performance and can map onto candidate characteristics such as race and gender. It also reports that candidates see cybervetting as a privacy violation that lowers procedural justice, whatever the hiring outcome.
For a talent leader this has a practical meaning. Unstructured searching adds exposure and employer-brand cost, and it adds little signal.
What to research, and what to leave out
A simple filter works: research only what a candidate would reasonably expect an employer to review, and only what predicts performance in the role.
Generally appropriate
- Professional profiles and public business content, such as a LinkedIn profile or industry forum posts.
- Verifiable credentials, licenses and employment history.
- Background checks that are relevant to the position and run through the proper legal process.
Leave out
- Personal social accounts that are not professional or business-related.
- Content posted when the candidate was a minor, and very old content. One university policy, for example, excludes content more than 7 years old or posted before age 18.
- Anything obtained by asking for login credentials. The same policy forbids this, and so do the best practices in a 2025 guide for California employers.
- Medical and genetic information. The EEOC's background-check guidance says employers should not seek genetic information, including family medical history, except in rare circumstances, and should not ask medical questions before a conditional job offer.
Lack of a public social account should never count against a candidate. Many strong candidates keep a low profile, and treating that as a signal introduces its own bias.
Sourcing ethics: consent is the dividing line
Sourcing is where most teams have the least written guidance. A practical test is whether the candidate would be surprised by how you found them and what you did with their data.
Reading a public profile and sending one personal, accurate message is normal recruiting. Harvesting profiles in bulk into a private database and messaging them automatically is different. A 2026 industry article describes this as scale without consent. People post profiles to be found professionally, not to be copied into third-party systems they cannot opt out of.
The same article cites a 2026 benchmark finding that 46% of sourced hires now come from candidates already in the company's database, up from 26% in 2021. Its suggested alternatives are practical:
- Start with existing talent pools and alumni.
- Build opt-in pipelines through talent communities, careers-site sign-ups and events.
- Use referrals.
- Pay for official platform licenses instead of scraping.
The ethics paper cited above adds two further sourcing rules. Do not pose as press or a directory publisher to gather information, because this casts a cloud over the firm's ethical behavior. Do not eliminate prospects during sourcing because of age, sex, race or other biases.
For executive search, one more issue applies: back-channel references. Informal calls to people the candidate did not name raise problems of consent, bias from unstructured conversations and confidentiality with the candidate's current employer. If your executive recruiting team uses them, the policy should say when, with whose permission and how they are recorded.
The practical risks of informal searching
The core legal problem with informal online research is simple. A recruiter who looks at a profile may see age, religion, national origin, pregnancy, marital status or disability. If a decision is later challenged, it is hard to prove that information was not considered.
The University of Southern Mississippi's hiring toolkit lists these protected categories and recommends using an unaffiliated third party that reports only job-relevant factors. A SHRM article from 2018 says social-media screening can lead to Title VII and state-law discrimination claims. Title VII is the federal law that bars employment discrimination based on race, color, religion, sex and national origin. SHRM's answer is an "ethics wall": a third party that makes no hiring decisions and passes on only job-related data after an offer, with protected characteristics removed.
A 2025 guide from Public Counsel makes the same point: an employer may be liable if a court finds that protected information influenced the decision.
Keeping research consistent and fair
Consistency is the main control against bias. These practices recur across the sources:
- Apply one search standard to every candidate for the same role. Uniform procedures are what make the process defensible. A law-firm guide from 2024 advises applying online research uniformly and recording what was found and how it affected the decision.
- Search late. Wait until after an initial assessment based on the resume and interviews, and focus on professional platforms.
- Separate the researcher from the decision-maker. The researcher removes protected-status content before anything reaches the hiring manager. The University of Iowa policy treats this as a structural control, not a matter of individual discretion. The same policy says social media must not be the only sourcing method because it can narrow the applicant pool.
- Test the criteria, not only the people. The EEOC's guidance on employment tests says procedures must be validated for the position and purpose, scores must not be adjusted by protected class, and an equally effective alternative with less adverse impact should be adopted when one exists.
- Watch for disparate impact in background rules. A policy that excludes people with certain criminal records can be unlawful if it significantly disadvantages a protected group and does not accurately predict who will be a responsible, reliable or safe employee.
For diversity recruiting leaders, item 5 matters most. A rule that looks neutral can still screen out a protected group.
The US rules, and what has changed
There is no single US rulebook for candidate research. Several layers apply at once.
Federal anti-discrimination law. The EEOC's background-check guidance states that whenever background information is used in an employment decision, however it was obtained, the employer must comply with federal anti-discrimination laws and apply the same standards to everyone. Note that BakerHostetler reported on November 11, 2026 that the EEOC has rescinded its 2012 background-check guidance. The underlying statutes remain. Treat the guidance as a statement of sound practice and confirm the current position with employment counsel.
The Fair Credit Reporting Act (FCRA). When a background check runs through a consumer reporting company, the EEOC's guidance says the employer must give a stand-alone written notice, obtain written authorization and certify to the reporting company that it notified the applicant and will not misuse the information. If the result leads to an adverse decision, a 2026 compliance playbook describes the usual sequence: a pre-adverse notice with the report copy and rights summary, a reasonable waiting period (commonly at least 5 business days), then a final notice naming the reporting company and the candidate's right to dispute.
State and local law. The same BakerHostetler piece notes that at least 24 states and localities require individualized inquiries before disqualifying someone based on criminal history. Violating their procedural requirements is itself a violation, even without disparate impact. Examples from a 2026 state-by-state guide:
- Philadelphia's amended Fair Criminal Record Screening Standards took effect January 6, 2026. The misdemeanor lookback fell from 7 to 4 years, and applicants get a 10-day written response window before final decisions.
- Washington's Fair Chance Act amendments took effect July 1, 2026 for employers with 15 or more employees. They bar criminal-history questions on applications and before a conditional offer, and require written notice, a waiting period and individualized assessments.
AI in sourcing and screening. Illinois HB 3773, effective January 1, 2026, amends the Illinois Human Rights Act to prohibit AI use with a discriminatory effect in recruitment, hiring and other employment decisions. According to TechRepublic, the CFPB's 2024 guidance treating algorithmic employment scores as FCRA consumer reports was rescinded in 2025, leaving courts to decide whether AI screening scores fall under the FCRA. The same report notes NYC's Local Law 144 already requires bias audits. An academic review from 2026 observes that federal enforcement against algorithmic discrimination has been limited by an April 2025 executive order, while state and local regulation continues to expand.
For a multi-state employer, the practical result is that the strictest jurisdiction in which you hire often sets the standard for the whole process. Counsel should confirm which rules apply to each requisition.
Building the policy
The sources agree on the components. A written policy should contain the following.
- Scope and relevance. State which information is relevant, for which roles and at what stage. A 2026 screening guide recommends tiers that define job-relevance.
- Who researches. Name the people permitted to run searches, and keep them separate from decision-makers.
- Notice. Tell candidates up front. The University of Iowa requires notice in the job posting that job-related internet content may be reviewed. If something found may affect the evaluation, the candidate sees it and can respond before any decision. Content and response go in the recruitment file.
- FCRA workflow. Include the notice, written authorization and standardized adverse-action steps described above.
- Documentation and retention. The University of Illinois recruitment guidelines, effective July 1, 2026, require interview notes, evaluations, interview questions and reference checks to be kept in the applicant tracking system, every applicant to receive a disposition code, and materials to be retained for three years after the calendar year they were created. Your retention period should come from your counsel and the laws of the states where you hire.
- Vendor oversight. A vendor-compliance guide suggests mapping hiring policies to vendor touchpoints such as data privacy, EEO, background screening and incident reporting. Pass or fail items include whether the vendor collects written candidate authorization before ordering a report. Keep the contract, data processing agreement and security certifications such as SOC 2 or ISO 27001 in each vendor record.
- Training. Train recruiters and hiring managers on what is appropriate and why, and review the policy on a regular schedule.
When the same questions come from different teams, the policy needs examples as well as rules. Use real scenarios: a candidate's public post about a medical condition, a reference call to a former colleague, an AI tool that ranks sourced profiles.
Where training and peer review help
Policy documents set the standard. Teams meet it through training and through seeing how other large functions handle the same cases. Recruiter training is part of every source above. IX Academy offers on-demand courses from 350 USD, live online courses from 750 USD and Team Intact courses from 3000 USD, all one-off prices as published on our site. For executive search teams, our ESIX community is built around the questions in this guide, and our DSIX community serves diversity recruiting leaders. Membership pricing is not published. Guest speaker sessions are described on the IX Communities guest speakers page.

To discuss membership for your talent or recruiting function, start at the IX Communities membership page.
Our perspective
Candidate research ethics is a policy question before it is a legal one. Most problems we would expect in a large recruiting function come from inconsistency: different recruiters, different searches, different records.
The benchmark for a policy is what peers actually do. A written standard that was tested against how other large corporate talent functions handle sourcing, references and background checks is easier to defend than one written alone.
This is the reason our communities are limited to in-house corporate recruiting leaders. Our environment is vendor-free and confidential, so members can describe what worked and what did not without a sales conversation attached. Benchmarking in our groups is peer-sourced. Heads of talent acquisition, executive recruiting and diversity recruiting can compare policies, notice language and vendor oversight practices with colleagues who carry the same responsibility.
We have operated for nearly three decades. Our view is that ethical research is a function standard, owned by the head of the function and checked against peers.
Simon
FAQ
Is it ethical to Google a candidate?
It can be, when it is limited to professional public information, applied to every candidate for the role, done after an initial assessment and disclosed in the policy. The researcher should remove protected-status content before the hiring manager sees results, and the findings should be documented.
Do we have to tell candidates we are researching them?
Notice is a recommended practice in the sources above, and a legal requirement for FCRA background checks, which need a stand-alone written notice and written authorization. Stating the practice in the job posting, as the University of Iowa does, also builds trust. Confirm state and local requirements with counsel.
What is the line between sourcing and evaluating?
Sourcing is finding people through public information and contacting them on your own initiative. Evaluating them can trigger consumer-report obligations, particularly when third-party tools produce scores or reports about them. This distinction matters most when you use AI sourcing tools, so ask counsel where each tool falls.
Who should run candidate research?
Someone separate from the hiring decision-maker. SHRM's guidance describes a third party behind an ethics wall. An internal HR or recruiting team member with no role in the decision, trained and following a written protocol, serves the same purpose.
How should we handle criminal history?
Follow the state and local rules where the role is located. Many require individualized assessments, post-offer timing and specific notices. Policies that exclude people automatically can create disparate impact liability if they are not job-related.
Sources
- Cybervetting in hiring: the hunt for moral performances (OUP Blog, 2021)
- Is cybervetting valuable? (Cambridge, 2022)
- Ethics of Recruiting and Selection (2017)
- The Use of Social Media in Recruitment (University of Iowa HR)
- A Guide for California Nonprofits and Small Businesses (Public Counsel, 2025)
- Background Checks: What Employers Need to Know (EEOC)
- Employment Tests and Selection Procedures (EEOC)
- Screening Candidates' Social Media May Lead to TMI, Discrimination Claims (SHRM, 2018)
- Hiring Toolkit for Staff Positions (University of Southern Mississippi)
- Legal and Ethical Considerations of Googling Job Applicants (Steadfast Employment, 2024)
- EEOC Rescinds Background Check Guidance (BakerHostetler, 2026)
- Background Check Compliance: A 2026 Employer Playbook
- Background Check Laws by State (2026)
- Quarterly Compliance Update: Summer 2025 (Cisive)
- The law is catching up to your AI screening tools (TechRepublic, 2026)
- An Examination of Regulatory Approaches to Artificial Intelligence (LMU Law Review, 2026)
- Compliant Employment Background Screening Best Practices (2026)
- Recruitment and Hiring Guidelines Effective July 1, 2026 (University of Illinois)
- How To Monitor Vendor Compliance Policies (2026)
- Ethical Candidate Sourcing: Build Pipelines Without Scraping (2026)
