Findings and credibility templates
Credibility Assessment Template & Example (Workplace Investigation)
Last updated 2026-07-30
Credibility is the part of an investigation most likely to be attacked, because it is the part most often asserted rather than reasoned. "I found the complainant more credible than the respondent" is a conclusion with the analysis deleted. It tells a reviewer nothing about why, and it cannot be checked.
A defensible credibility determination does three things: it identifies the specific factual conflict being resolved, it applies objective factors to that conflict, and it writes down which factor drove the outcome. Everything else — including how the witness came across — is secondary and, in the case of demeanor, actively dangerous to rely on.
This template applies the factors most commonly used by professional investigators: inherent plausibility, corroboration, internal consistency, consistency with contemporaneous records, motive to fabricate, and demeanor with appropriate caution. The worked example below shows what the finished reasoning looks like, including the harder case where the assessment does not resolve the conflict.
Jurisdiction
Written primarily against US federal standards — EEOC guidance, NLRB decisions and Title VII case law — and against general investigative practice. State law, sector rules (public employers, education, regulated industries) and collective agreements impose additional requirements this page does not cover. Where a page borrows a concept from another jurisdiction it says so explicitly. Outside the US, treat the structure as transferable and the legal references as not.
Credibility Assessment template
Free Word download, no email required. Or open it in Professional Drafter — it lands in your template library, ready to draft against.
When to use this document
- Any material factual conflict where the outcome turns on whose account is accepted.
- One-to-one incidents with no witnesses and no documentary corroboration.
- Where a party's account has changed between the complaint and the interview.
Worked example
Fictional scenario, continuing the dispatch-team matter used across this library. Two conflicts: one resolved on corroboration and contemporaneous records, one left unresolved. The second is included deliberately — worked examples that only show clean resolutions teach the wrong lesson.
Conflict 1 — whether the comments continued after being asked to stop
Disputed fact: whether the complainant asked the respondent to stop commenting on her appearance in early April 2026, and whether comments followed.
Competing accounts: The complainant states she asked him directly, in the corridor outside dispatch, in the first week of April. The respondent does not recall any such request, but does not deny that one was made.
Inherent plausibility: Both accounts are plausible. A brief verbal request of this kind is easily forgotten by the person receiving it, particularly if they did not regard the comments as unwelcome — which is consistent with the respondent's characterization of them as compliments.
Corroboration: Witness A recalls the complainant telling her, around the same period, that she had 'told him to knock it off'. This corroborates that the complainant made the request and reported doing so at the time. It does not independently establish the respondent heard it.
Contemporaneous records: The message at Appendix 4 is dated 14 March, which is before the request as the complainant describes it, so it does not assist on sequence. However, the complainant's own text to Witness A of 8 April refers to having 'said something to him yesterday' — a contemporaneous record created before any complaint was made.
Motive: I identified no motive to fabricate on either side. The complainant did not seek any outcome that would benefit from this specific allegation. The respondent's motive to deny is the ordinary one facing anyone under investigation, and I have not treated it as evidence against him.
Demeanor: Both parties were composed. I place no weight on this.
Determination: I accept, on the preponderance of the evidence, that the request was made in early April and that the respondent received it. The determination rests on the contemporaneous text of 8 April and Witness A's independent recollection, not on any assessment of who presented better. The respondent's failure to recall the request is not treated as untruthful — it is equally consistent with his account that he did not regard the comments as a problem.
Conflict 2 — the corridor incident of approximately 17 April (unresolved)
Disputed fact: whether the respondent physically blocked the complainant's path in the corridor outside the dispatch office.
Competing accounts: The complainant describes him stepping into her path and remaining there for several seconds while she attempted to pass. The respondent denies the incident occurred and cannot recall being in that area at that time.
Inherent plausibility: Both accounts are plausible. The corridor is a normal route between the dispatch office and the loading bay, so his presence there would be unremarkable, and equally his absence would be unremarkable.
Corroboration: None either way. Witness A recalls seeing the complainant appearing distressed near the dispatch office on a Friday in mid-April, which is consistent with the complainant's timeline but does not speak to what caused the distress or whether the respondent was present.
Contemporaneous records: The corridor CCTV camera has been out of service since January 2026 (Facilities, 18 June). Door access logs for that corridor are not maintained. No message or note from either party refers to the incident before the complaint of 29 May.
Internal consistency: The complainant's account has been consistent across three tellings, and she volunteered that she could not recall whether anyone else was in the corridor — a concession against her own interest that I regard as a marker of candour rather than of accuracy. The respondent's account is a general denial, which is internally consistent but not capable of being tested.
Motive: None identified on either side.
Determination: The conflict is not resolved. The factors that decided Conflict 1 — contemporaneous records and independent corroboration — are entirely absent here, and the remaining factors point in neither direction. I have not carried my acceptance of the complainant's account on Allegations 1 and 2 across to this allegation: credibility accepted on one point does not establish an uncorroborated account on another, and reasoning that way would substitute a general impression for evidence. Allegation 3 is accordingly recorded as inconclusive, which is a statement about the evidence and not a finding that the incident did not occur.
Section-by-section guide
What each section is for, what a good one contains, and where they typically go wrong.
1.The conflict being resolved
State the disputed fact narrowly. Credibility is assessed conflict by conflict, not witness by witness — a witness may be reliable on one point and mistaken on another, and a global 'this person is credible' finding hides that.
2.Competing accounts
Each account set out neutrally, with the source and date of each version given.
3.Inherent plausibility
Is the account plausible on its face, given the setting, the timing and what else is known? Take care here: plausibility is a check on internal coherence, not a test of whether the conduct seems out of character. Reasoning from 'he isn't the type' is a well-documented route to error.
4.Corroboration
What independent evidence supports each account — witnesses, documents, messages, access records. Note the strength of the corroboration, not merely its existence: a witness who recalls the complainant being upset corroborates distress, not the incident.
5.Internal consistency
Whether each account has stayed consistent across tellings. Weigh changes carefully — peripheral detail commonly shifts and minor inconsistency is normal in truthful accounts. What matters is movement on the central facts, and whether a change was volunteered or extracted.
6.Consistency with contemporaneous records
How each account sits against what was recorded at the time — messages, diary notes, rosters, access logs, emails. Contemporaneous material is usually the most reliable evidence available because it was created before anyone had a reason to shape it.
7.Motive to fabricate
Whether either party had a reason to lie, and how strong it is. Assess it symmetrically — a respondent facing dismissal has an obvious motive too, and considering only the complainant's motive is a recognizable form of bias.
8.Demeanor — recorded with caution
If you record demeanor at all, record the observation and then state that it is not being relied on as a primary factor. The evidence here is unusually strong and unusually ignored: the standard meta-analysis (Bond & DePaulo, 2006, pooling 206 documents and 24,483 judges) found people identify lies and truths correctly about 54% of the time — barely above chance, and notably worse at catching lies (47%) than at recognizing truths (61%). Nervousness, flat affect and inconsistent eye contact also vary with culture, neurodivergence and trauma response. A determination resting on demeanor is both the least accurate kind and the easiest to overturn.
9.Determination and reasoning
Which account is accepted, on the preponderance of the evidence, and — most importantly — which factors drove that. Name them. If the factors do not resolve the conflict, say so and record that the point is unresolved rather than defaulting to whoever seemed more convincing.
Common mistakes
The failure modes that show up most often when this document is reviewed later.
Making a global credibility finding about a person
Credibility attaches to accounts of specific facts, not to people. A witness can be reliable about one incident and mistaken about another, and a blanket finding conceals that.
Relying on demeanor
Nervousness, flat affect and avoided eye contact are poor indicators of deception and vary with culture, neurodivergence and trauma response. Determinations resting on demeanor are the least defensible kind.
Treating minor inconsistency as dishonesty
Truthful accounts routinely vary in peripheral detail; word-perfect consistency across tellings is itself sometimes a warning sign. Weigh movement on central facts, not on incidentals.
Considering motive on only one side
A respondent facing dismissal has an obvious motive to deny. Examining the complainant's motive while ignoring the respondent's is a recognizable asymmetry that a reviewer will spot immediately.
Carrying a credibility finding across allegations
Accepting someone's account on one incident does not establish an uncorroborated account on another. That reasoning replaces evidence with general impression.
Forcing a resolution where the evidence does not support one
An honest 'unresolved' is defensible. A determination manufactured because the investigator felt obliged to decide is not, and it is usually visible in the thinness of the stated reasoning.
Questions
- Both parties seem honest and the accounts are irreconcilable. Do I have to pick one?
- No. If plausibility, corroboration, consistency, contemporaneous records and motive genuinely do not separate the accounts, record the conflict as unresolved and make an inconclusive finding on the allegation that depends on it. An honest unresolved conflict is defensible; a determination manufactured because you felt obliged to decide is usually visible in how thin the stated reasoning is.
- Can I assess the credibility of someone I never interviewed?
- Be very careful. You can weigh a document or a statement on its face, but a credibility determination about a person whose account you obtained second-hand — from another investigator's notes, or from what a party says they said — is difficult to defend. If their account matters, interview them. If you cannot, say so and explain what that means for the finding.
- Can an investigator rely on demeanor to decide who is telling the truth?
- It is unwise, and the research is clear about why. Bond and DePaulo's 2006 meta-analysis of 206 documents and 24,483 judges found average accuracy of about 54% — close to chance, and worse for detecting lies (47%) than truths (61%). Demeanor also varies with culture, neurodivergence and trauma response. If you record it, state expressly that it is not a primary factor and identify what the determination actually rests on.
- What if there are no witnesses and no documents — a pure 'he said, she said'?
- Work the remaining factors: internal consistency, plausibility, motive, and any indirect contemporaneous material such as a message to a friend or a change in work patterns. If none of them resolve it, record an inconclusive finding with the reasoning. That is a legitimate and defensible outcome.
The blank template
Exactly what is in the Word download, so you can see the structure before deciding whether you want it.
The conflict being resolved
Competing accounts
Inherent plausibility
Corroboration
Internal consistency
Consistency with contemporaneous records
Motive to fabricate
Demeanor — recorded with caution
Determination and reasoning
Download this template
Free Word download, no email required. Or open it in Professional Drafter — it lands in your template library, ready to draft against.
Sources
Related templates
Findings of Fact Template: Substantiated, Not Substantiated, Inconclusive
A findings-of-fact worksheet that ties each allegation to the evidence relied on, the standard applied, and a clear substantiated / not substantiated / inconclusive outcome.
Workplace Investigation Report Template & Example
A complete workplace investigation report template with a filled-in example report, plus a section-by-section guide covering scope, evidence, credibility, standard of proof and findings.
Investigation Interview Questions for the Accused Employee
A respondent interview template covering notice, the right to respond to each allegation, and the questioning sequence that makes a finding defensible.
Workplace Bullying Investigation Report Template & Example
A bullying investigation report template distinguishing bullying from legitimate management action, with a note on why US bullying complaints are policy matters rather than statutory ones.
Inconclusive Investigation Outcome: Honesty Test, Next Steps, Monitoring
Inconclusive means the evidence did not permit a determination. The checks to run before you write it, what the employer can do next, and how to monitor.
This page is a documentation aid, not legal advice. Workplace investigation requirements vary by jurisdiction, sector and policy, and a template cannot tell you what your matter requires. The worked example describes a fictional scenario.