⚡ Promptolis Original · Decisions & Reasoning

🔮 Pre-Mortem Analysis Tool

Imagine the project failed — now explain why. The Gary Klein technique that catches 30% more failure modes than traditional risk analysis, because 'why did this fail' unlocks different reasoning than 'what could go wrong.'

⏱️ 4 min to run 🤖 ~60 seconds in Claude 🗓️ Updated 2026-04-19

Why this is epic

Traditional risk lists miss the failure modes that matter because they're framed as 'what could go wrong' — a vague, optional question. Pre-mortem frames it as 'it failed; tell me why,' which forces specific, plausible story-telling and surfaces 30% more real risks.

Gives you the 6 specific failure categories (assumptions wrong, capability gap, coordination failure, market change, stakeholder block, self-sabotage) and runs each independently — most pre-mortems clump them and miss half.

Produces ranked failure probabilities with early-warning indicators for each — so you're not just aware of risks, you have a dashboard of what to monitor in real-time.

The prompt

Promptolis Original · Copy-ready
<role> You are a decision-science practitioner trained in Gary Klein's pre-mortem methodology. You have run 300+ pre-mortem sessions across product, strategy, and personal decisions. You will force the user to imagine the decision failed, then explain in specific detail why. You are allowed to surface uncomfortable failure modes the user would prefer to avoid thinking about. </role> <principles> 1. Past-tense framing ('it failed because...') surfaces more specific, plausible risks than future-tense ('what could go wrong'). 2. Generate at least 5 failure scenarios across 6 categories: assumptions wrong, capability gap, coordination failure, market/environment change, stakeholder block, self-sabotage. 3. Rank scenarios by probability AND impact. High-P high-I first. 4. The output that matters: early-warning indicators for each scenario. 5. Include at least one uncomfortable scenario the user didn't want to name. 6. Re-read schedule matters. Pre-mortem reviewed at 1/3/6 months earns its keep. </principles> <input> <the-decision-or-project>{what you're considering}</the-decision-or-project> <success-definition>{what 'this worked' looks like concretely}</success-definition> <time-horizon>{when will you know}</time-horizon> <investment>{time, money, relational cost being committed}</investment> <current-risks-you-see>{risks you already know about}</current-risks-you-see> <your-track-record>{have you done similar before, how did it go}</your-track-record> </input> <output-format> # Pre-Mortem: [Project / decision] ## The Frame Imagine it's [time horizon] from now and this failed. The setup. ## 6 Failure Scenarios One per category: 1. Assumptions wrong 2. Capability gap 3. Coordination failure 4. Market / environment change 5. Stakeholder block 6. Self-sabotage Each in past tense, specific. ## Scenario Ranking By probability × impact. Top 3 highlighted. ## Early-Warning Indicators For each top-3 scenario: what would you see 3-6 months in if this scenario were materializing? ## The Uncomfortable Scenario The one you didn't want to name but should. ## Pre-Commit Mitigations What you can change NOW to lower the top-3 probabilities. ## The Review Schedule When to re-read this pre-mortem. ## The Kill-Switch Under what specific conditions you'd abandon the project. ## Key Takeaways 4 bullets. </output-format> <auto-intake> If incomplete: ask for the decision, success definition, time horizon, investment, known risks, track record. </auto-intake> Now, run the pre-mortem:
🚀 Open in ChatGPT ✨ Open in Claude 💎 Open in Gemini

Example: input → output

Here's how this prompt actually performs. Real input below, real output from Claude Opus 4.

📝 Input
<the-decision-or-project>Launching a paid subscription tier ($15/mo) on our free productivity app. Current: 80k free users, 0 paid. Target: 5,000 paid in 6 months = $75k MRR.</the-decision-or-project>
<success-definition>5,000 paid subscribers at 6 months with <15% monthly churn. Net positive contribution after ad/referral spend.</success-definition>
<time-horizon>6 months.</time-horizon>
<investment>$40k engineering (3 months dev), $60k launch marketing budget, 2 team members primary focus. Total ~$150k with opportunity cost.</investment>
<current-risks-you-see>Low conversion from free (we have no baseline). Features behind paywall might not be 'must-have' enough. Tight marketing budget.</current-risks-you-see>
<your-track-record>Founder has launched 2 previous freemium products — one worked (~8% conversion), one failed (<1% conversion). Current team has not launched paid tier before.</your-track-record>
🤖 Output

Pre-Mortem: Paid subscription tier on free productivity app, 6-month horizon

A pre-mortem analysis is a structured failure-imagination exercise — 'imagine it failed, tell me why' — that surfaces ~30% more real risk factors than traditional prospective risk analysis. Based on Gary Klein's research and 300+ practitioner sessions, past-tense specific narratives surface risks that future-tense speculation misses because the brain generates failure explanations more fluently than predictions. Your specific profile — prior 1-for-2 freemium track record, tight budget, 6-month horizon — makes this a high-value pre-mortem; you have real data for the base rate but not for this team.

The Frame

It's 6 months from today. The paid tier has 1,200 subscribers (not 5,000). Churn is 22%/month. You're at $18k MRR, net negative after marketing spend. The team is exhausted, the board is skeptical, and you're debating whether to kill the tier, raise the price, or pivot the business model. How did this happen?

6 Failure Scenarios

1. Assumptions wrong — 'The free users weren't the market for the paid tier.'

Your free user base optimized for 'free.' They signed up because it was free; they'd churn at any price. The paid tier's actual market was a different segment (teams, power users) you didn't build the product for. You spent 6 months converting the wrong audience.

2. Capability gap — 'The team couldn't run a subscription business.'

Engineering built the paywall; nobody ran customer success, payment failure recovery, or churn fighting. Cards expired silently. Support tickets for billing piled up. The operational muscle of running subscriptions was entirely absent.

3. Coordination failure — 'Marketing and product were misaligned on the pitch.'

Marketing sold the paid tier as 'unlock pro features.' Product positioned it as 'support the creators.' Users arrived expecting one thing and got another. Conversion stalled because the message kept shifting across channels.

4. Market / environment change — 'A competitor made it free.'

In month 3, a VC-funded competitor made the exact features you're charging for fully free, bundled into their existing offering. Your paid tier suddenly had to justify itself against free. Momentum collapsed.

5. Stakeholder block — 'Existing users revolted.'

Free users experienced the paywall as a bait-and-switch. Public blog posts, Reddit threads, and App Store reviews tanked your brand. You reversed 3 of the paywall features under pressure, neutering the value proposition.

6. Self-sabotage — 'The founder couldn't commit to the pivot.'

Part of you didn't fully believe in charging. Pricing page updates got delayed. Upgrade CTAs got softened. The team sensed founder ambivalence and mirrored it. The launch was 'quiet' instead of loud, and nobody noticed the paid tier existed.

Scenario Ranking (Probability × Impact)

Top 3 (high probability, high impact):

1. #1 Wrong audience — 45% probability. Your 1-for-2 track record and lack of paid-tier validation support this being dominant.

2. #6 Self-sabotage / half-commitment — 30% probability. Founders who have shipped free products for years often under-commit to paid. Watch this.

3. #2 Capability gap — 25% probability. No one on your team has run subscriptions. Operational muscle takes 3-6 months to build.

Lower tier:

4. #3 Coordination failure — 15%

5. #5 Stakeholder block — 12% (moderate but recoverable)

6. #4 Competitor making it free — 8% (real but external; limited mitigations)

Early-Warning Indicators (Monitor Weekly)

Scenario #1 (Wrong audience):

  • Week 2-4 of launch: conversion rate among top-10%-engaged free users is <3%
  • No clear segment with 10%+ conversion emerges after 1,000 upgrade prompts
  • Upgraders don't retain differently from a 'willing-to-pay' cohort you identified pre-launch

Scenario #6 (Self-sabotage):

  • Pricing page changes get delayed past planned dates
  • The team starts discussing 'maybe we should make X feature free again'
  • You find yourself avoiding looking at the upgrade dashboard

Scenario #2 (Capability gap):

  • Billing support tickets >20/week with no dedicated owner
  • Passive churn (expired cards) accounting for >30% of churn
  • No clear retention playbook for month 2-3 users

The Uncomfortable Scenario

#6 — Self-sabotage / half-commitment. Your input says the team has not launched paid before, and your prior success was your earlier product, not this one. There's a specific under-identified risk: you built the free app as a passion project, and charging for it feels like breaking a social contract with your users. If you're even 20% ambivalent, the team will sense it and the launch will be muted.

This is the scenario you should name explicitly to the team this week.

Pre-Commit Mitigations (Do Before Launch)

1. Run a paid-intent survey of top-engaged free users BEFORE building. If <15% say they'd pay, you've already confirmed scenario #1 — don't launch.

2. Hire or appoint one person (existing team member ok) to own 'customer success' before launch. Billing, retention, payment failure recovery.

3. Write the public launch narrative THIS WEEK and have it approved by founder + marketing + product + support. One version. Shared.

4. Identify the founder's commitment level honestly. If <8/10, delay launch and work on the commitment first.

5. Set price-drop / feature-reversal rules in advance so stakeholder-block (#5) pressure doesn't destabilize you in-flight.

The Review Schedule

  • Week 2 post-launch: Check scenario #1 and #6 indicators
  • Month 1: Full re-read of pre-mortem + compare to reality
  • Month 3: Re-read. Update probabilities based on evidence.
  • Month 6: Compare prediction vs. outcome. What did the pre-mortem get right / wrong? Calibrate for next time.

The Kill-Switch

If by month 3 ANY of the following are true, seriously evaluate killing or radically restructuring:

  • <1,000 paid subscribers (20% of goal)
  • Monthly churn >25%
  • CAC payback >18 months
  • Team morale collapsed AND conversion <3%

The kill-switch is not 'we'll decide at month 6' — because by month 6 you'll have another $50k of sunk cost. Decide at month 3 with clean criteria.

Key Takeaways

  • Your highest-probability failure mode is 'wrong audience' (#1) at ~45%. Run the paid-intent survey BEFORE building, not after.
  • Self-sabotage / half-commitment (#6) is the uncomfortable one. Have the founder commitment conversation with the team this week.
  • Build customer-success operational muscle BEFORE the paywall ships. Billing + retention are first-order requirements, not afterthoughts.
  • Month 3 kill-switch with clean criteria. Don't let this run to month 6 on momentum alone.

Common use cases

  • Before a major product launch
  • Before signing a multi-year contract / partnership
  • Before a significant hire at senior level
  • Before committing to a strategic pivot
  • Before a large marketing / campaign spend
  • Before a major life decision (relocation, career pivot)
  • Any decision where failure cost > $100k or >6 months of time

Best AI model for this

Claude Sonnet 4.5 or Opus 4. Good pre-mortem requires generative scenario reasoning. Mid-tier and above.

Pro tips

  • Do the pre-mortem BEFORE you publicly commit to the decision. Once committed, your brain defends the plan and pre-mortem loses power.
  • Run it with a team member who disagrees with you, or at minimum someone skeptical. Echo-chambers produce weak pre-mortems.
  • Write the failure story in past tense ('the launch failed because...'). Past-tense framing is more specific than hypothetical.
  • Generate MORE failure stories than feels comfortable. 5 scenarios is a good floor; 10 is better.
  • The most useful output is the early-warning indicators, not the scenarios themselves. Monitor those weekly.
  • After the decision, re-read the pre-mortem at 1 month, 3 months, 6 months. The prophetic ones earn their keep.

Customization tips

  • Run this WITH your co-founder or team — pre-mortems done solo tend to generate 2-3 scenarios; team-run pre-mortems generate 6-8.
  • Write each scenario in PAST tense before you read any of them. Past-tense forces specific storytelling.
  • Keep the pre-mortem document open and visible during the project. Reviewing it at month 1 with the team is one of the highest-ROI hours you'll have.
  • If a scenario comes up that you dismiss too quickly ('that won't happen'), mark it and revisit — dismissal is often a signal.
  • After the project, calibrate: which scenarios did you predict? Which did you miss? The pre-mortem is also a learning tool for your own reasoning.

Variants

Product Launch Mode

Specifically for pre-launch risk analysis. Weighted toward market, timing, and adoption failures.

Personal Life Decision

For individual life decisions (move, career change, relationship). Weighted toward emotional / relational failure modes.

Team / Hire Mode

For senior hires or team restructures. Weighted toward coordination, culture, and capability failures.

Frequently asked questions

How do I use the Pre-Mortem Analysis Tool prompt?

Open the prompt page, click 'Copy prompt', paste it into ChatGPT, Claude, or Gemini, and replace the placeholders in curly braces with your real input. The prompt is also launchable directly in each model with one click.

Which AI model works best with Pre-Mortem Analysis Tool?

Claude Sonnet 4.5 or Opus 4. Good pre-mortem requires generative scenario reasoning. Mid-tier and above.

Can I customize the Pre-Mortem Analysis Tool prompt for my use case?

Yes — every Promptolis Original is designed to be customized. Key levers: Do the pre-mortem BEFORE you publicly commit to the decision. Once committed, your brain defends the plan and pre-mortem loses power.; Run it with a team member who disagrees with you, or at minimum someone skeptical. Echo-chambers produce weak pre-mortems.

Explore more Originals

Hand-crafted 2026-grade prompts that actually change how you work.

← All Promptolis Originals
P

Curated by

Every Promptolis Original is hand-crafted and reviewed before publishing — built from scratch for 2026-grade LLMs.

Last reviewed on · About Promptolis