Hello! Welcome back to our module on Strategic Foresight & Failure Analysis.
In our last lesson, we introduced the powerful mental model of inversion, where you work backward from an imagined failure to identify potential catastrophic failure modes. You practiced this by conducting an "inversion pass" on a new AI feature, which helped you uncover systemic risks beyond simple bugs.
Today, we'll build directly on that foundation. We will move from the individual act of identifying failures to a structured, team-based process for analyzing them. Your learning outcome for this lesson is to conduct a pre-mortem analysis to proactively identify risks for a new feature launch. This is a critical skill for any architect looking to build robust and resilient systems.
1. From Inversion to Pre-Mortem: Why It Works
While inversion is a great personal mental tool, its real power in an organization comes from a formal exercise called a pre-mortem. A post-mortem analyzes a project after it has already failed. A pre-mortem is a prospective autopsy: you imagine the project has failed before you even start, to figure out what could kill it and prevent those outcomes.
Why is this simple reframing so effective? It's about psychology. In most project kickoffs, the team is encouraged to be optimistic. Questioning the plan can be seen as negative or disloyal. The pre-mortem flips this on its head.
To understand the psychological power of this technique, let's hear from Nobel laureate Daniel Kahneman, whose work you may know from Thinking, Fast and Slow.
Nobel laureate Daniel Kahneman - Premortem to eliminate thinking biases.
In this short clip, Daniel Kahneman explains why the pre-mortem is his favorite technique for 'debiasing' decisions. He highlights how it overcomes organizational optimism bias and legitimizes dissent.
Watch from 00:17 to 02:24. Focus on his explanation of how the pre-mortem 'legitimizes dissent' and rewards people for finding flaws in a plan, which is normally a socially risky thing to do in a team.
By making failure the starting assumption, the pre-mortem creates psychological safety. It channels the team's collective experience and anxieties into a productive, forward-looking analysis instead of letting them fester as unspoken doubts.
2. How to Conduct a Pre-Mortem Analysis
A pre-mortem is not just an unstructured complaint session; it's a facilitated process. While there are many ways to run one, they generally follow a clear set of steps.
The video below gives a great overview of a standard, six-step process.
What is a Pre-mortem? ...and how do you run one?
The video 'What is a Pre-mortem?' from Online PM Courses provides a concise, step-by-step guide to facilitating a pre-mortem session.
Please watch from 02:53 to 05:33. Pay attention to the six distinct steps he outlines: Preparation, Imagine a fiasco, Generate reasons, Consolidate, Review, and Determine next steps.
This structured approach is similar to how an engineer might use a Failure Mode and Effects Analysis (FMEA) to systematically analyze potential failures in a mechanical system. Here, we're applying that same rigor to a product system.
3. A Framework for Failure: The Four Ways Products Die
To make your pre-mortem effective, you need to ensure you're looking for risks in all the right places. It's easy for a team to get stuck on technical issues alone, but products often fail for reasons that have nothing to do with code.
The following article provides an excellent framework by categorizing product failures into four key areas. It's filled with real-world examples that bring these risks to life.
The Premortem: Your Product's Autopsy Before Launch
The article 'The Premortem: Your Product's Autopsy Before Launch' from Eleganthack is a superb guide to the practical application of this technique. The section on the four failure categories is particularly valuable.
Please read the section titled 'The Four Ways Products Die'. Focus on understanding each of the four categories: Technical, Market, Ethical, and Regulatory/Environmental. Note the examples given for each, as they are highly relevant to modern product design.
As a UI/UX designer, the Ethical Failures section is especially crucial. Notice how products like Nextdoor's crime reporting and Robinhood's gamification worked exactly as designed from a technical and engagement perspective, but created significant unintended harm. A pre-mortem using this framework can help you spot those risks before you ship.
4. From Analysis to Action: Making the Pre-Mortem Stick
The biggest pitfall of a pre-mortem is treating it as a one-time intellectual exercise. The output must be a living document that leads to concrete action. The goal is not just to list problems, but to actively mitigate them.
Shreyas Doshi, a former product leader at Stripe and Twitter, developed a modified pre-mortem script designed to create lasting impact. He introduces a helpful vocabulary to create psychological safety and a clear process for creating an action plan.
How to Use Pre-mortems to Prevent Problems, Blunders, and ...
This article by Shreyas Doshi offers a refined, actionable script for running a pre-mortem. It introduces a powerful vocabulary for discussing risks and emphasizes the creation of a follow-up action plan.
Please read the sections 'Enter the metaphors: Tigers, Paper Tigers, Elephants' and the section that begins 'So, what’s next? It’s the most important part...'. Focus on what a 'Tiger' is, and understand the components of the 'hypothetical example' action plan: the problem (Tiger/Elephant), the DRI (Directly Responsible Individual), and the mitigating actions.
The concept of assigning a DRI (Directly Responsible Individual) is critical. It transforms a vague concern into an owned action item, ensuring that someone is accountable for investigating the risk and implementing the mitigation plan.
5. Practical Exercise: Conducting a Pre-Mortem
Now, let's put this all together. You will step into the role of a facilitator and conduct a solo pre-mortem analysis for the same feature we discussed in the last lesson.
Scenario:
You are the lead designer for "TaskFlow," a B2B SaaS project management tool. One month before the scheduled launch of the new premium feature, "AI Smart Scheduler," you are running a pre-mortem.
- Reminder of what it does: The feature uses AI to automatically analyze project tasks and assign them to team members based on their skills, workload, and availability.
Your Task:
-
Set the Stage (Imagine Failure): Your starting point is the classic pre-mortem prompt: "It is six months after launch. The AI Smart Scheduler has been a complete disaster. We've had to disable it, our support lines are overwhelmed, and two of our largest clients have canceled their contracts because of it."
-
Brainstorm Failure Modes (Categorize): Based on this premise, brainstorm at least eight potential reasons for this failure. Use the four categories from the Eleganthack article as your guide, listing at least two potential failure modes for each:
- Technical/Operational (e.g., scalability, reliability, data issues)
- Market/Business (e.g., pricing, competition, user adoption)
- Social/Ethical (e.g., bias, user harm, unintended consequences)
- Environmental/Regulatory (e.g., legal challenges, data privacy laws)
-
Prioritize and Plan (Identify Tigers & DRIs):
- Review your list of eight failure modes. Identify the top three "Tigers"—the risks with the highest probability and highest potential impact that could truly kill the feature.
- For each of these three Tigers, create an action plan item using the format from Shreyas Doshi's article:
- Problem (The Tiger): Briefly state the risk.
- Proposed Mitigating Action: What is a concrete first step the team could take now (before launch) to reduce this risk? Be specific (e.g., "Conduct bias audit on training data," not "Be less biased").
- Suggested DRI: Who on the team would be the logical person to own this action? (e.g., "Lead Engineer," "UX Researcher," "Product Manager," "Legal Counsel").
Take about 20 minutes to complete this analysis. This exercise will move you from simply identifying problems to structuring, prioritizing, and planning for them.
Conclusion
In this lesson, you have formalized the practice of inversion into a structured pre-mortem analysis. This is a powerful tool for moving from reactive problem-solving to proactive risk mitigation.
Key Takeaways:
- A pre-mortem is a structured exercise that assumes project failure to uncover risks. It creates psychological safety for teams to voice concerns.
- A comprehensive analysis should consider risks across multiple domains: Technical, Market, Ethical, and Regulatory.
- The goal is not just to list risks but to create an action plan that assigns a Directly Responsible Individual (DRI) to each critical issue ("Tiger").
- This process transforms vague anxiety into a concrete, prioritized plan, making your product and team more resilient.
Next Lesson Preview:
In this lesson and the last, we've focused on identifying and mitigating specific, individual failure modes. In the next lesson, we will zoom out to analyze a system's overall character in response to stress. We will explore the concepts of fragility, robustness, and antifragility from Nassim Nicholas Taleb's work, asking not just "how does this break?" but "how does the system as a whole react to the unexpected?"