Making Better Bets: How Decision Making Applies in Workplace Investigations

I just finished reading "Thinking in Bets: Making Smarter Decisions When You Don't Have All the Facts" by Annie Duke. What does a former poker champion know about making good decisions in business? It turns out, quite a bit. Annie recommends a solid strategy by establishing key processes that help you make good decisions despite our tendency to focus on what she calls "resulting" - the focus on outcomes that can negatively impact our decision-making process when we put too much confidence in bad decisions because of good outcomes, or change good decision-making processes as a result of poor outcomes.

You might be wondering how this relates to workplace investigations. Our role as investigators is to uncover facts and ultimately make decisions on the balance of probabilities about what most likely occurred. A big part of Annie's strategy focuses on probabilities. Assessing credibility in workplace investigations is what Annie would call a "bet," which should always be based on probabilities rather than certainties.

The Investigation Process Framework

My investigation process has multiple steps but generally follows the same flow, with some steps repeated depending on complexity and clarity:

1. Complete complaint intake - Detail each individual complaint if more than one. Obtain witnesses or other parties potentially involved.

2. Inform respondent(s) of the complaint

3. Schedule and conduct initial interviews while obtaining evidence (documents, recordings, etc.). Typically in the order of complainant, witnesses, and respondents.

4. Conduct follow-up interviews based on new information from secondary interviews (being considerate of scope creep)

5. Detail summary of facts and findings - usually peer reviewed by a colleague.

6. Complete the final report

   
   

Making Better Bets: Five Key Strategies for Workplace Investigations

1. Focus on Process Over Outcomes

Just as poker players can make good decisions and still lose hands due to chance, investigators can follow proper procedures and still face criticism if stakeholders disagree with findings. We shouldn’t have a vested interest in the outcome if we are truly neutral. The key is maintaining consistent, thorough processes regardless of whether your conclusions are popular. Document your methodology, ensure comprehensive evidence collection, and perform interviews with empathy and neutrality. This creates a solid lens through which to process information while recognizing and removing biases. Ideally every investigator starts the case without a preconception of guilt. 

2. Consider All Data Regardless of Source

All information needs consideration, but the critical questions are: Can it be validated? Is it authentic? Determine source credibility by examining whether information remained consistent over time, has it been corroborated by other witnesses, and whether evidence supports allegations based on claimed locations and timelines. Don't dismiss uncomfortable evidence simply because it complicates a narrative. Again be aware of any preferences or biases that might arise throughout the investigation.

3. Form "Decision Groups" to Combat Confirmation Bias

Annie Duke emphasizes accountability groups that challenge your thinking. In workplace investigations, build systematic review processes with colleagues or supervisors before reaching final credibility determinations. Present preliminary findings to trusted colleagues who can ask probing questions like: "What evidence contradicts your current theory?" or "What assumptions are you making about the complainant's motivations?" This external perspective helps identify when you might be cherry-picking supporting evidence while dismissing contradictory information. Challenging yourself to question your decisions from the alternative perspective can also help you to ensure you are able to document your rationale for the decisions you have made. This can also come in the form of a peer review at the end of an investigation prior to finalization.

4. Assign Confidence Levels to Each Finding

Rather than making a complete binary "credible vs. not credible" determination, assign probability percentages to each conclusion. For each allegation, assign confidence levels: "I'm 75% confident harassment occurred based on corroborating witness statements and text message evidence, but 25% uncertain due to gaps in the timeline." This forces acknowledgment of uncertainty, prevents overconfidence, creates better documentation for reasoning, and helps identify which areas need additional investigation. Ultimately our client is looking for a decision of credible or not credible based on whether there a greater than 50% chance the alleged conduct occurred, so looking at decisions in these terms can better help you get to your final conclusion. 

5. Conduct Pre-mortem Analysis Before Finalizing Reports

Before completing your final report, conduct systematic reviews by asking: "If this finding is later proven incorrect, what would be the most likely reasons?" This might reveal overlooked evidence, untested alternative explanations, or witness credibility issues not fully explored. For example, if concluding harassment didn't occur, consider scenarios where it did happen but wasn't adequately documented, or where power dynamics prevented full disclosure.

The Ultimate Goal

Making better bets in workplace investigations means assessing credibility based on comprehensive fact-gathering, individual story consistency, evidence quality, and relevant information that could influence outcomes. The trick is maintaining a solid investigative lens that removes biases, ensures comprehensive data collection, and enables logical, rational decisions while reaching findings.

By applying Annie Duke's betting framework to workplace investigations, we move beyond "resulting" and build systematic processes that acknowledge uncertainty, seek diverse perspectives, and test conclusions before finalizing them. This approach doesn't guarantee perfect outcomes, but it significantly improves the quality and defensibility of our investigative decisions.

Recommended reading: "Thinking in Bets: Making Smarter Decisions When you Don't Have All the Facts" Annie Duke https://www.annieduke.com/books/

Or if you want a brief overview, listen to Gene Killian talk about it on his podcast Station 4 Negotiation https://directory.libsyn.com/episode/index/show/d28a3d93-b827-4bd7-a91b-99711b518a8f/id/33174357