top of page
Search

HR Investigation Interview Questions: What to Ask Employees, Witnesses, and Complainants

Updated: 8 hours ago

The right question, asked the right way, is the whole game. In a workplace investigation, your questions either draw out an accurate, uncontaminated account or they lead the witness, taint the evidence, and hand opposing counsel an argument. This is a working bank of HR investigation interview questions for complainants, witnesses, and subjects, plus the questions you should never ask. It's part of our full guide to workplace investigations; for the end-to-end method, see how to conduct a workplace investigation interview.


The Short Answer: The best HR investigation questions are open, neutral, and non-leading. Start by asking the person for a complete free narrative account in their own words, then move into focused follow-up questions about people, actions, locations, timing, communications, and supporting evidence.


Table of Contents


Asking open HR investigation interview questions.
Asking open HR investigation interview questions.

One rule before the questions: how you ask beats what you ask

You can have a perfect list and still ruin an interview by delivering it wrong. Three rules make any question work:


  • Keep it open. Favor questions that invite a story, not a yes/no.

  • Ask one thing at a time. Compound questions confuse the record and let people answer only the easy half.

  • No multiple-choice questions. Multiple-choice questions narrow what you are looking for and limit an interviewee's response.

  • Don't signal the answer. Leading questions put your words in their mouths—and that's exactly what contaminates a statement.


This is the heart of science-based interviewing: let the person talk, listen actively, and follow their account instead of your theory or your supervisor's theory.


Start here: the free-recall prompt

Every science-based interview starts with breadth, not detail. Before any specific questions, invite the person to give their whole account of the incident, in their own words, and then stop talking, no interruptions:

"Tell me, in your own words and in as much detail as you can, everything you remember about [the event/situation].

Let it run uninterrupted. This could be the most valuable answer you'll get all day—the entire picture, in their words, free of your assumptions. As they talk, you're doing two things at once: learning what they actually know and noting what you're curious about. Those notes added to your pre-interview preparation become your map.


Only after the account is finished do you move from breadth to depth—going back to drill, or funnel, into the specifics, one interview objective at a time. Breadth first, depth second: get the whole story before you dig into any one part of it. You never know; you may end up with an unknown, unknown.


Opening questions (set the tone)

  • "Do you understand why we're meeting today?"

  • "Before we start, do you have any questions about the process?"

  • "Is there anything you need to be comfortable — water, a break, anything?"


These aren't filler. They establish fairness and cooperation—and how people are treated in the room is how your whole function gets judged.


Questions for the complainant

  • "Walk me through what happened, from the beginning."

  • "When and where did this happen?"

  • "Who else was present or nearby?"

  • "What exactly was said or done? As close to the words as you can remember."

  • "How did you respond at the time?"

  • "Has anything like this happened before or since?"

  • "Did you tell anyone about it? Who, and when?"

  • "Is there anything — messages, emails, photos, documents — that relates to this?"

  • "How has this affected your work?"

  • "Is there anything I haven't asked that I should have?"


Questions for witnesses

  • "What did you personally see or hear?"

  • "Where were you when this happened?"

  • "Tell me exactly what you observed—not what you concluded."

  • "Did you see how each person reacted?"

  • "Is what you're describing something you saw yourself or something you heard about? From whom?"

  • "Has anyone spoken with you about this incident before today?"

  • "Is there anyone else who might have seen or heard something?"


That "saw it vs. heard it" distinction is critical—it's how you keep rumor out of your findings.


Questions for the subject / respondent

Give the subject a fair, non-accusatory chance to account for themselves before you disclose what you know:


  • "Tell me about your working relationship with [complainant]."

  • "Walk me through [date/event] as you remember it."

  • "What happened from your perspective?"

  • "Is there anything about that day that stood out to you?"

  • "Are you aware of any reason someone might raise a concern about this?"

  • "Is there anything you'd like to add or clarify?"


Hold your evidence until they've committed to an account—that's Strategic Use of Evidence, and it's what turns a vague denial into a testable one without pressure.


Harassment investigation questions

Harassment complaints run on the same method—free recall first for breadth, then depth—but there are facts you have to pin down here that other matters don't require. Some of this overlaps with the complainant set above, and it should; the core questions don't change, but the emphasis does. Start with the free-recall prompt, let the account run, then work through the core set of harassment investigation questions:


  • "Tell me what happened."

  • "What exactly was said or done?" —as close to the actual words as they can recall

  • "When did this happen?"

  • "Where did this happen?"

  • "Describe for me who else was present?"

  • "Was this the only time this has occurred?"

  • "What happened immediately afterward?"

  • "Who did you tell?" (Outcry witness)

  • "What records, messages, emails, or other communications might exist?"


Then add the probes specific to harassment matters:


  • Frequency and severity. "How often did this happen?" "Did it change over time—how?"

  • Effect on the work. "How has this affected your ability to do your job?"

  • How they responded at the time. "How did you respond?" "Did you say anything to them about it?" Ask it flat and neutral—never, "Why didn't you just tell him to stop?"

  • Notice. "Did anyone in a management or supervisory role know about this?" Whether the organization knew or should have known is often the key issue.

  • Prior similar conduct. "Are you aware of anyone else who has experienced this?" "Have you witnessed something similar?"

  • Retaliation. "Has anything changed at work since you reported this incident?" Ask it in the interview and again at the follow-up—retaliation frequently shows up after the investigation opens, not before.


Three cautions that matter more in harassment cases than anywhere else:


Delay is normal — don't investigate it like a defect. People wait for reasons that have nothing to do with truthfulness. If timing is relevant, explore it without blame through evocation-based questions like "Walk me through what you thought about doing after it happened" or "what were you thinking when that occurred?"


Gather facts, not legal conclusions. Don't ask a complainant or witness whether the conduct was "harassment," "hostile," or "unwelcome" as a label. Those are conclusions for the decision-maker and counsel to apply to your findings. Your job is what was said, what was done, how often, and who saw it. (See how to separate findings, conclusions, and recommendations in the report.)


Protect the reporting party. Harassment interviews are where source exposure does the most damage. Hold your evidence back whether technology or statements, and if you decide to disclose it in a form that doesn't identify who told you—Strategic Use of Evidence (SUE) lets you test a subject's account without handing them the name of the person who reported it or information you have.


Probing and follow-up interview questions

Once you have a strong free narrative, you switch from breadth to depth. Using funnel questioning, go back through the account and drill into the objectives the case turns on—actions, people, locations, capabilities, opportunities, and timelines—expanding each answer without ever leading. You're following their account and satisfying your curiosity from the initial free narrative, not planting your own version.


Start with open prompts to reopen a topic or a funnel:

  • "Tell me more about that."

  • "What happened next?"

  • "Help me understand…"

  • "You mentioned XYZ — walk me through that."


Then get into the specifics:

  • Actions: "Walk me through exactly what was said." "What did you do right after that?"

  • People: "Who else was there?" "Who else would know about this?" "Who did you tell?"

  • Locations: "Where were you when this happened?" "Where was everyone else?"

  • Timeline: "What happened just before that?" "About what time was it?" "In what order did those things happen?"

  • Basis of knowledge: "How do you know that?" "How did you learn about it?"


Keep every probe neutral (not leading) and one at a time. The moment a follow-up question hands them the answer ("So he threatened you, right?"), you've stopped gathering information and started contaminating it.


Open, memory-friendly prompts like these come straight from cognitive interviewing, and they pull far more accurate detail and more of it than rapid-fire closed questions (this hurts rapport and is still taught in some interviewing training).


Closing questions

  • "Is there anything else I should know?"

  • "Is there anyone else I should talk to?"

  • "Is there anything you'd like to correct or add to what you told me?"

  • "If I have follow-up questions, is it alright if I contact you again?"


Questions to avoid

These quietly wreck investigations:


  • Leading: "He threatened you, didn't he?" (You just wrote their answer.)

  • Compound: "Did you see it and did you report it?" (Which half are they answering?)

  • Assumptive: "How long has this been going on?" (Assumes it happened.)

  • Accusatory "why": "Why didn't you just say no?" (Feels like blame; use evocation.)

  • Opinion-seeking: "Do you think he's guilty?" (You want observations, not verdicts.)

  • Demeanor traps: anything based on reading nerves, tells, cues, or eye contact—that kind of lie detection doesn't work, and it injects bias into your findings.


Write down the answer, not your conclusion

Record what was said, as closely as possible, and keep your interpretations out of the notes. Assess credibility later through consistency, corroboration, and detail—not a gut read in the moment or a "lie detection" cue that sounds more like hocus pocus.


What established guidance recommends asking

Our question set lines up with what the leading guides recommend — and adds the discipline of keeping every question open and non-leading. The most detailed public question banks, found online, come from the NALC manager's guide to investigating harassment, which gives separate question sets for the complainant, the alleged subject, and witnesses, and FordHarrison's HR-attorney guidance, which stresses "foundation of knowledge" probes and the legal elements of a harassment claim. Across these and other guides (Heffernan, Cravath), the recurring moves are consistent.


  • Establish the basis of every account—"How do you know that? Did you see it or hear about it?"

  • Always ask who else knows and what evidence exists—other witnesses, messages, documents, or footage.

  • For harassment, probe frequency, severity, and pattern—was it welcome, how often, how severe, and did it affect the work?

  • Surface any prior similar allegations.

  • Close by confirming completeness — "Is there anything else I should know?"


HR investigation interview questions Frequently Asked Questions

What questions should you ask in an HR investigation?

Open, neutral, non-leading ones—start with a free-recall prompt, then probe with "tell me more," "what happened next," and "help me understand." Ask complainants, witnesses, and subjects each in their own sequence.


Leading, compound, multiple choice, embedded probes, rapid-fire closed-ended, assumptive, and accusatory "why" questions—and anything that relies on reading body language.


Fewer than you think. One strong free-recall prompt plus targeted follow-ups on objectives beats a long checklist read start to finish.

 It depends on your organization's policy, applicable law, and any union or collective-bargaining agreement—check those first. As a general practice, investigators don't share the questions ahead of time, since it lets the account be pre-scripted. But many settings require advance notice of the allegations and the right to representation or a support person. Whatever your rules require, still give the subject a fair chance to give their full account before you disclose any evidence.

HR should begin with open, neutral questions that allow the employee to explain what happened in their own words. Good questions focus on what the person saw, heard, did, or experienced; when and where events occurred; who else may have relevant information; and what documents, messages, emails, or other evidence may exist. Follow-up questions should clarify details without suggesting the answer.

Start by asking the employee to describe the situation from beginning to end in their own words. Then ask focused follow-up questions about the people involved, timing, locations, communications, actions, and any supporting evidence. If the employee is responding to an allegation, give them a fair opportunity to explain their account, identify witnesses, and provide information that may support or contradict the allegation.

Ask witnesses what they personally observed, heard, or experienced. Separate firsthand knowledge from assumptions or information they learned from someone else. Useful follow-up questions include when and where the event occurred, who was present, what happened immediately before and after, whether there were prior similar events, and whether any records or communications may help verify the account.

Avoid leading, loaded, compound, rapid-fire, or accusatory questions that suggest the answer you expect. Questions such as “Why did you harass her?” assume the allegation is already true. A more neutral approach would be, “Tell me about your interaction with her that afternoon.” Investigators should also avoid questions based on assumptions, behavioral stereotypes, or attempts to judge truthfulness from nervousness, eye contact, posture, microexpressions, or other unreliable behavioral cues. No one testifies to this stuff.

Begin with a neutral explanation of the interview and allow the employee to give a full account before moving into more specific questions. Listen carefully, clarify important details, and explore alternative explanations rather than treating the allegation as established fact. Relevant evidence should be planned for in the preparation stage strategically so the investigator can compare the employee’s account with known information without unnecessarily contaminating the statement.

Do not respond to resistance by becoming more confrontational or with more adaptive behaviors. Explain the purpose of the interview, maintain a professional and respectful tone, and use open-ended questions and active listening to encourage the employee to talk. Acknowledge concerns without agreeing with unverified claims. Rapport, patience, and careful follow-up questions often produce more information than pressure or repeated accusations.

A strong opening question is broad and gives the person room to provide an uninterrupted account. Examples include, “Tell me everything you remember about what happened,” or “Walk me through what occurred from the beginning.” The investigator can then move from broad questions to more focused questions based on the information provided.

Investigators should red team and actively consider more than one possible explanation for the evidence. Before and during interviews, identify assumptions, look for information that could challenge the working theory, and ask questions that could support or disprove competing explanations. The goal is not to confirm the initial allegation; it is to determine what the available information supports.

An HR investigative interview should be designed to gather accurate and complete information rather than obtain a predetermined admission. "Interrogation" is a legal term applied to U.S. law enforcement questioning a suspect who is in custody. Science-based investigative interviewing relies on rapport, open-ended questions, active listening, careful follow-up, and thoughtful evidence use. A confession or admission may occur, but it should be the result of information gathering rather than the sole objective of the interview.




Comments

bottom of page