Our team of specialist workplace investigators has made thousands of findings across more than 1,500 investigations. This article draws on their collective experience, along with insights from case law, to provide practical guidance on how to make findings that withstand scrutiny.
KEY TAKEAWAYS:
Making findings is often the most demanding stage of an investigation and the point where investigations are most at risk of going wrong.
- Sound reasoning requires the investigator to make their thinking – how they have carefully considered and weighed the evidence before them – transparent and logical, so others can follow how they have reached their conclusions.
- Findings must meet the civil standard of proof, that is, the balance of probabilities. For serious allegations, the Briginshaw principle requires the fact-finder to be genuinely persuaded about what occurred, not merely conclude that one version is marginally more likely than another.
- Not all evidence carries equal weight – the more the reasoning is anchored in objective factors, the more defensible it will be.
The evidence has been gathered, now comes the stage that many investigators find the most demanding – making findings of fact.
This stage requires investigators to step back from the information they have collected, weigh it carefully, and reach conclusions they can clearly explain and defend. It is challenging work that weighs heavily on investigators, and it is where investigations are most at risk of going wrong.
Findings are the foundation upon which critical management decisions are made. Decisions like dismissing or formally disciplining an employee or telling a complainant their version of events could not be substantiated. A flawed finding can damage careers, relationships and reputations, and expose an employer to unfair dismissal claims or other legal challenges.
The stakes go beyond legal risk. If the investigation process is not well managed, including at the findings stage, there is a risk of inadvertently causing psychological harm to participants. Weak reasoning, unexplained outcomes, and a lack of follow‑through with participants don’t just undermine the investigation, they also create psychosocial hazards.
Keep the purpose of the report in mind
The investigation report exists to put the decision-maker in the best position possible to decide what action to take. The investigator is not, and should not be, the decision-maker.
The report must adequately convey the investigation process, the relevant evidence, the reasoning and the findings so that the decision-maker can act soundly. The omission of key information, such as clear reasoning, does not just weaken the findings, it can cause the entire decision-making process to fail.
It is best practice to write the report on the assumption that the complainant and respondent may read it at some stage, and that it may be subject to scrutiny in legal proceedings. The report should stand alone as a document, readable and understandable without reference to other materials.
A note on timing: Making findings and writing the report is time-intensive, focused work usually requiring hours of uninterrupted concentration. Allow sufficient time to draft, review, and refine. Once the report is drafted, sleep on it. Reviewing it with fresh eyes the next day will often reveal gaps in reasoning that were not visible when writing. This is especially important for finely balanced findings.
What reasoning is (and what it is not)
At its core, reasoning is the disciplined process of analysing the evidence and weighing it in a structured way to reach a conclusion that can be clearly explained and defended. It’s like showing the workings in a maths problem.
It is not a retelling of the facts or the allegations with a conclusion at the end that offers no justification.
Good reasoning requires the investigator to consider: which facts are actually proven; how the evidence supports or contradicts itself; which interpretations genuinely fit the evidence; and whether overall the evidence meets the required standard of proof.
Logical and clearly articulated reasoning is not just important for the end reader, or whoever reviews the work, but also for the investigator themselves. Committing reasoning to writing is an important step in analysing the evidence and identifying gaps and inconsistencies.
Sound reasoning examines all relevant evidence; it does not start with a conclusion and select only the evidence in support of that conclusion. This is one of the most serious errors an investigator can make, and it is also one of the hardest to detect in your own work.
Practical tips:
- Scan your draft findings for unsupported statements and words that sound authoritative, such as ‘clearly’, which push the reader toward a conclusion rather than letting the evidence do the work.
- Where a conclusion is sound, the evidence should demonstrate it without the investigator needing to assert it. For example:
| Unsupported assertion | Sound reasoning |
| The respondent clearly intended to stay at the impacted person’s house. | The respondent’s Uber account shows his home address was not entered as a third stop. The investigator found his explanation, that this was a mistake because he was drunk, to be unconvincing, given he was capable of correctly entering the impacted person’s address into the application. In the circumstances, it is reasonable to conclude that the respondent intended to stay at the impacted person’s house. |
- Before drafting any findings, organise the evidence. Build a timeline of the relevant events. List the facts that are uncontested versus those in dispute. Map out where the accounts align and where they conflict. This helps to make patterns clear before the reasoning process begins.
Standard of proof
Findings must meet the civil standard of proof: that is, the balance of probabilities. This means the investigator must be satisfied that it is more probable (or likely) than not that the alleged conduct occurred. If the standard of proof is not met, the allegation cannot be substantiated.
The Briginshaw v Briginshaw (1938) 60 CLR 336 standard adds an important layer: the more serious the allegation, the higher the evidentiary standard required. As the High Court established in Briginshaw, reasonable satisfaction that something has occurred cannot be attained without taking into account the nature and seriousness of the alleged act, the gravity of the consequences, and the inherent unlikelihood of the event happening. An allegation of serious misconduct therefore demands more compelling evidence than a minor policy breach.
In the Federal Court’s judgment in Lehrmann v Network Ten Pty Limited [2024] FCA 369, the court was required to determine, on the balance of probabilities, whether a rape had occurred. This decision reinforced the Briginshaw principles and offers contemporary practical guidance for workplace investigators. The court emphasised that the standard requires more than a mechanical comparison of probabilities. A finding on the balance of probabilities requires the fact-finder to reach a genuine state of persuasion about what occurred, not merely conclude that one version is marginally more likely than another.
In practice, this means placing the competing accounts alongside the available evidence and asking: which version is better supported by credible evidence? Not more evidence, but more credible evidence. It also means being precise about what the evidence establishes versus what it merely points towards. These are different things and conflating them is a common analytical error.
A four-step approach to making findings
Step 1 — Consider all relevant evidence
Ask yourself: does this evidence help me determine whether the alleged conduct occurred? If the answer is no, leave it out. Irrelevant evidence clutters reasoning and can weaken an otherwise sound finding. Equally, the summary of evidence must address both evidence that supports the finding and evidence that does not support it.
For example, a witness’s observations about an impacted person’s clothing or general personality traits are unlikely to assist in determining whether a specific incident of sexual harassment occurred. By contrast, contemporaneous documents that corroborate witness evidence (such as the Uber transaction record set out in the example above), are likely to be directly relevant and ought to be given significant weight due to their contemporaneous and objective qualities.
Step 2 — Disregard opinion and speculation
Anchor findings firmly in factual evidence. Evidence that represents a witness’s view, interpretation, or assumption about events, rather than a factual account of what they observed, should be disregarded. For example, in a sexual harassment investigation, witness evidence that the respondent was ‘definitely not like that’ because he was married with children, is a personal opinion and not evidence of what occurred.
Step 3 — Assess the quality of evidence and how much weight it deserves
Not all evidence carries equal weight. Think of the available credibility factors on a scale, from the most objective to the least. The more the reasoning is anchored in objective factors, the more defensible it will be.
Most objective – and most reliable evidence
Corroborating evidence is the gold standard. That is documentary, physical, or witness evidence that independently supports one account over another.
The Lehrmann judgment reinforced what experienced investigators already know: in cases involving competing accounts, a contemporaneous document, such as a text message, email, transaction record (such as an Uber receipt), or file note created at or near the time of the event, carries significant weight precisely because it is not subject to the frailties of human memory.
Opportunity for knowledge: That is, did the witness have the opportunity to obtain the information, for example to observe the conduct in question? Being present does not automatically mean a witness was in a position to see or hear what is alleged.
Consistency: Is the account consistent across multiple sources, that is, throughout the interview, any written statement, and any contemporaneous communications. The presence of inconsistencies in a person’s account do not, however, prove dishonesty per se. For example, trauma can fragment memories, and memory can change over time. Where inconsistencies arise, put them to the interviewee and ask for an explanation before drawing any conclusions.
Less objective — use with care
Motive and bias: Is there any evidence that a witness has a reason to modify or omit information? Take for example a sexual harassment investigation where the respondent asserts that the impacted person had fabricated her evidence to obtain a pay rise. The investigator will need to carefully consider this assertion in the context of all evidence and decide how much weight to give it. In circumstances where there is no evidence to corroborate the assertion, and the impacted person voluntarily provided detailed evidence, including information unfavourable to her, the investigator may decide to place little weight on this assertion. A potential motive is a factor to weigh, not a conclusion in itself.
Plausibility: Does the account make sense in the context of everything else the evidence shows? Be careful — plausibility assessments can be highly subjective. Test reasoning against objective evidence before relying on it.
Avoid or treat with significant caution
Demeanour of an interviewee during interview is unreliable evidence. Investigators are not trained to interpret physical or emotional responses, and demeanour is an unreliable indicator of truthfulness. Demeanour can provide the investigator with helpful signals on when to seek further information on a particular issue, but beyond this, it is generally unhelpful.
Reputation evidence and prior complaint history are to be treated with great caution. Save for exceptional circumstances, each matter is to be assessed on its own merits.
Step 4 – Apply common sense
Consider the likelihood of the conduct in question occurring. The more inherently unlikely a version of events, the stronger the evidence required. Even where a witness appeared honest and open, consider whether they had a strong or weak recollection; or whether their recollection may have been impacted by the time elapsed, alcohol consumption, trauma, or any other relevant factor.
Also keep in mind that rejecting one part of a person’s evidence does not require rejecting all of it. Justice Lee in Lehrmann accepted aspects of both parties’ evidence while rejecting other parts. Be careful not to characterise every incorrect statement as untruthful. Often an equally plausible explanation is that the person simply misremembered.
More information
Q Workplace Solutions’ team of experienced and legally qualified investigators is trusted by public and private organisations, including ASX-listed companies and government agencies, to investigate complex and often highly sensitive allegations of employee wrongdoing. The team also undertakes reviews of organisations, divisions, or units, and provides training, coaching and external advisory support to internal investigators and teams.
Upcoming training | 22 April 2026
Master how to write clear and fair investigation reports with well-reasoned findings that stand up to scrutiny in this 22 April online workshop – Evidence to Outcome: Making Findings and Report Writing. Places are limited to ensure a highly interactive professional learning experience. Group bookings available. Register here.
