What Courts Look For When a Workplace Investigation Is Challenged

An employer finishes an investigation, reaches a conclusion, and takes action. Months later, the matter is in litigation or before the EEOC. The question the adjudicator is asking is not only whether the employer got it right — it's whether the investigation itself was conducted with enough care to be trusted. Courts and agencies have developed a fairly consistent set of markers they look for, and understanding those markers helps practitioners build investigations that hold up before they're ever challenged.

The Reasonable Investigation Standard

The most developed body of law on investigation quality runs through the Faragher/Ellerth affirmative defense, which lets an employer escape vicarious liability for supervisor harassment if it exercised reasonable care to prevent and promptly correct the harassing behavior. See Faragher v. City of Boca Raton, 524 U.S. 775 (1998) and Burlington Industries v. Ellerth, 524 U.S. 742 (1998). Because the defense turns on the adequacy of the employer's corrective response, the quality of the investigation becomes legally decisive — which is why the richest case law on what a "good investigation" looks like sits in the harassment context.

"Reasonable care" in the correction prong means the investigation itself has to be reasonable — not perfect, but methodologically sound. Courts applying this standard have looked consistently at scope, documentation, impartiality, and whether the investigator followed up on leads rather than accepting the first plausible answer.

The EEOC's guidance on harassment reinforces this framework, instructing that investigations should be prompt, thorough, and impartial. These aren't aspirational adjectives — they function as evaluation criteria in agency proceedings.

A note on scope. The standards in this article originate in harassment law, but the markers themselves are not confined to it. See "The Markers Travel Beyond Harassment," below, for how the same criteria are imported into ordinary misconduct discipline, discrimination, and just-cause arbitration through different legal doctrines.

What Courts Look For in the Record

When a workplace investigation ends up in discovery, courts examine several specific elements:

Scope and follow-through. Investigators who interview every available witness and pursue documentary evidence — emails, access logs, scheduling records, prior complaints — are in a much better position than those who speak only to the principals and close the file. The flip side is instructive: in EEOC v. Management Hospitality of Racine, Inc., 666 F.3d 422 (7th Cir. 2012), the court treated prompt investigation as a hallmark of reasonable corrective action and let a jury verdict against the employer stand where managers ignored repeated complaints and a meaningful investigation began only after an outside investigator started asking questions. Investigations that fail to interview obvious witnesses or ignore readily available documentary evidence are exactly what courts hold against employers.

Timeline documentation. When investigators took each action — when they received the complaint, when they notified HR leadership, when they conducted each interview, when they concluded their work — matters procedurally and substantively. A timeline that can be reconstructed from contemporaneous records is far more credible than one reassembled from memory during a deposition. Management Hospitality underscores the point from the other direction: a months-long gap between complaint and investigation was itself evidence that the response was not reasonably prompt. Date-stamped interview notes, logged communications, and clearly dated draft and final reports all contribute to a coherent timeline that demonstrates the investigation moved appropriately and systematically.

Citation practices in reports. A findings section that attributes every material fact to a specific source — "Witness A stated in her October 14 interview that…" — is demonstrably more defensible than one that synthesizes evidence without attribution. Courts and agencies want to trace conclusions back to evidence. Reports that state conclusions as if they emerged from nowhere give opposing counsel significant leverage. The citation practice that matters most is consistency: every factual assertion should be anchored to something in the record.

Findings language. The way conclusions are written signals whether the investigator applied the right analytical framework. The EEOC and best-practice commentators consistently recommend using the preponderance standard — "more likely than not" — as the stated basis for findings. Findings written as certainties ("It is established that…") or as impermissible hedges ("It is possible that…") both create problems: the former overreaches the standard of proof, the latter suggests the investigator did not actually reach a conclusion. Language like "The evidence supports a finding that, more likely than not…" is both accurate and defensible.

Where the Analysis Gets Complicated

Two areas generate recurring difficulty in practice.

First, investigator impartiality is assessed on facts, not job titles. An HR professional who has had prior documented conflict with a respondent, who reports directly to someone with a stake in the outcome, or who has made comments suggesting a pre-formed conclusion creates a documented impartiality problem. The SHRM guidance on workplace investigations recommends a formal assessment of potential conflicts before any investigator is assigned. The adequacy of the employer's response is what carries the day: in Knabe v. Boury Corp., 114 F.3d 407 (3d Cir. 1997) — decided a year before Faragher and Ellerth, under the earlier "prompt and adequate remedial action" standard — the Third Circuit held that what matters is whether the employer's response was reasonably calculated to prevent future harassment, not whether the investigation was flawless. The practical lesson for impartiality is that a response a reviewer can see was objective and untainted is far more durable than one a respondent or complainant can credibly attack as predetermined.

Second, adequacy of documentation during the investigation itself — not just in the final report — is increasingly scrutinized. Notes taken during witness interviews, preserved in a form close to contemporaneous, carry more evidentiary weight than polished summaries drafted days later. Investigators who record decisions made during the investigation — why a particular lead was not pursued, why a particular witness was not interviewed, how a scheduling conflict affected timing — create a record that explains the investigation's shape. Gaps in this real-time record tend to be filled in by opposing counsel's narrative.

The Markers Travel Beyond Harassment

The framework above comes from harassment law, and a reader could be forgiven for assuming Faragher/Ellerth governs every workplace investigation. It does not — but the same markers (promptness, thoroughness, impartiality, evidence-based conclusions) are imported into other contexts through different legal doctrines:

• Discrimination and retaliation generally. Under the McDonnell Douglas pretext framework, a one-sided or sloppy investigation becomes circumstantial evidence that the stated reason for an adverse action was a cover for a prohibited motive. The investigation's quality is doing evidentiary work even though the affirmative defense is nowhere in sight.
• Ordinary misconduct discipline (the "honest belief" doctrine). Several circuits shield an employer from liability if it held an honestly-formed, reasonably-grounded belief in the misconduct it disciplined for — but that protection collapses when the underlying investigation was inadequate. In Fisher v. Airgas USA, LLC (6th Cir. 2024), the court held the doctrine did not protect an employer that terminated an employee over a positive drug test without investigating the employee's specific, raised explanation that a legal hemp product may have caused a false positive. This was a drug-policy termination, not a harassment claim — yet the failure-to-investigate analysis is the same.
• Just cause (union / CBA arbitration). Labor arbitrators apply the Daugherty "seven tests of just cause," one of which requires a fair and objective investigation conducted before discipline is imposed. This standard predates Faragher/Ellerth and applies to any disciplinary investigation regardless of subject matter.

The takeaway: the markers are largely universal, but the legal consequence of failing them depends on the claim. A defensible investigation looks the same whether the underlying issue is harassment, theft, a safety violation, or attendance — what changes is which doctrine the inadequacy gets measured against.

What Good Practice Looks Like

Investigators who build defensible records do several things systematically: they document the complaint intake with specificity, maintain a running log of every investigative action with dates, take and preserve interview notes in a form tied to each witness and each session, cite every factual assertion in the final report to its source, apply a stated evidentiary standard in findings language, and document the basis for any procedural decisions that could later appear to have been convenient. The investigation doesn't have to be perfect — courts have consistently acknowledged that reasonable investigations may miss things. What they must be is traceable, methodical, and documented well enough that a reviewer can reconstruct what happened and why.

The standard, at bottom, is whether a reasonable fact-finder would look at the process and conclude the employer took the complaint seriously and investigated it with care.

---

This article is for general information and does not constitute legal advice. Cited authorities should be independently confirmed and shepardized before being relied on; case law and agency guidance change.

---

From Marshal

Marshal's audit chain logs every investigator action with a tamper-evident timestamp — creating exactly the kind of contemporaneous record courts look for when they reconstruct whether an investigation was conducted with care. Learn more →