Would I Start This Today? Test for Quitting

Making decisions explicit is the key to improving decision quality, especially in group settings. Annie Duke's frameworks help organizations identify blind spots, close feedback loops, and make better decisions even when outcomes are far in the future.

Improving Group Decision Making

The Discover-Discuss-Decide Framework

• People generally think meetings are for three things: discover, discuss, decide
• The only thing that should happen in meetings is the discussion part
• For discovery (gathering opinions):
  • Collect input independently and asynchronously before the meeting
  • Ask people to submit their thoughts in writing without seeing others' responses
  • For forecasts or estimates, get point estimates, lower bounds, and upper bounds
  • Have people provide rationales for their positions
• For discussion:
  • Focus on areas of disagreement rather than agreement
  • As a facilitator, reflect back what people say without offering your own opinion
  • Ask clarifying questions rather than challenging statements
• For decisions:
  • Make decisions outside the meeting, either by a single decision-maker or through independent voting
  • Avoid seeking "alignment" - it's unrealistic and creates coercive dynamics

Closing Feedback Loops

• "There is no such thing as a long feedback loop" - you choose how long your feedback loop is
• To shorten feedback loops:
  • Identify signals correlated with your desired long-term outcome
  • Track necessary but not sufficient conditions for success
  • Create intermediate checkpoints to evaluate progress
  • Example: For venture investments, don't wait 10 years for an exit - track series A funding, product-market fit, retention metrics

Premortems and Kill Criteria

• Premortems alone don't change behavior - they must be paired with pre-commitments
• Process:
  1. Imagine the project has failed and identify early warning signals
  2. For each signal, establish specific kill criteria
  3. Commit to concrete actions you'll take if those signals appear
• Example from sales:
  • Signal: Customer only wants to talk about price, not demo the product
  • Kill criteria: If they refuse a demo and only discuss price
  • Action: Stop pursuing the deal

Making Better Quitting Decisions

• By the time you're thinking about quitting, it's probably already past the time you should have quit
• Biases that prevent timely quitting:
  • Sunk cost fallacy - feeling you'll waste what you've already invested
  • Endowment effect - overvaluing things you've created
  • Identity concerns - fear of being seen as a failure
  • Uncertainty - waiting until you're 100% sure before quitting
• The cost of not quitting includes opportunity costs - you can't pursue new opportunities
• Ask "Would I start this today?" as a forecast when deciding whether to quit

Making Implicit Knowledge Explicit

• Our intuition is sometimes right and sometimes wrong
• Making implicit knowledge explicit allows you to:
  • Identify when intuition is wrong
  • Create structured decision processes
  • Evaluate the quality of past decisions
  • Learn and improve over time
• Even experts can be wrong about what factors are most important in their decisions

The power of these frameworks comes not from eliminating intuition but from making it explicit, testable, and improvable over time.

Lenny Rachitsky: When you look at companies that have read your book what do you find are the frameworks or tactics that really stick

Annie Duke: People generally think the purpose of a meeting is for three things discover discuss decide the only thing that's ever supposed to happen in a meeting is the discussion part

Lenny Rachitsky: Something that comes up in product a lot is this idea of premortem

Annie Duke: So a pre mortem is great only if you set up kill criteria commit to actions that you're gonna take if you see those signals