Your browser doesn't support the features required by impress.js, so you are presented with a simplified version of this presentation.

For the best experience please use the latest Chrome, Safari or Firefox browser.

Decision Making Mental Model

Philosophy and Principles

  • Good decisions are measured by the process, not the outcome: a good decision is one that was sized and assessed rationally, where we’ve made reasonable judgements on all options, and their probabilities of occurring, and then made the rational call based on that. We too often conflate the outcome (good or bad) with the quality of the decision. Outcomes are probabilistic.

  • What matters more in decisions? Analysis or Process? Both matter, but process matters more. Poor analysis and amazing process will by default weed out poor analysis.

  • The goal is not to make perfect decisions, but rather to make better decisions than average.

  • Improving the odds of making good decisions is the ability to distinguish between decisions within our competence, and those that aren’t.

  • Remember that often your role is to manage a coalition: “a manager is usually portrayed as a great decision maker — the scientific decision maker. She’s got her spreadsheet and she’s got her statistical tests and she’s going to weigh the various options. But in fact, real management is mostly about managing coalitions, maintaining support for a project so that it doesn’t evaporate”.

  • Decisions are part of a larger context of management. They have externalities, and offer opportunities:

    • Practicing/introducing accountability
    • Process of making the decision and rationale can be shared, offering your managers a chance to learn how to ‘read your mind’.
    • You can integrate people into the decision making process. Creating/evolving a ‘team mind’ with shared knowledge, goals and identity.
    • They affect and change other people’s strategies and goals.
  • Communication of decisions is highly undervalued, under-leveraged, and often poorly handled.

Decision Making Mental Model

Mental model: an explanation of someone’s thought process about how something works in the real world

Decision: a conclusion or resolution reached after consideration; a choice made between alternative courses of action in a situation of uncertainty.

Risk: something you can put a price on.

Uncertainty: risks that are hard to measure.


This second describes a high level model for how I think about decision making. Before using it, identify the following:

  • Does this decision affect/influence any goals? If so, what are they? List by priority and impact.

  • How big is the blast radius of this decision? List how many people it can possibly affect, and by how much.

  • List the people I can include in the decision making process to improve the chances of my decision being the best one.

  • List the people most likely to improve the truthiness of the decision making context.

  • Decide if this decision matters. Exclude decisions in the ‘zone of indifference’. If the effect is small, go with your gut. Else, continue.

What is the goal?

Identify the primary goal that will be used to measure the outcome of the decision.

What is the measurement?

Identify the correct measurement of the goal, ensure this measurement is/can be built.

This measurement will be used to assess the largest risk/reward payoff, and will also be used to continually monitor the decisions progress towards the desired outcome.


  • Determine what you know now
  • Compute the value of additional information
  • Measure where information value is high
  • The cost of measurement should be lower than the benefit of decision making

Philosophy of measurement

  • We should care about measurement because it informs key, but uncertain decisions.
  • Measurement is the methods and process for reducing uncertainty.
  • Soft, touchy-feely-sounding-things like ‘employee empowerment’, ‘creativity’ or ‘strategic alignment’ must have observable consequences if they matter at all.
  • Measurement: a quantitatively expressed reduction of uncertainty based on one or more observations.
  • Uncertainty: The existence of more than one possibility where the ‘true’ outcome is not known. Measurement of uncertainty is a set of probabilities.
  • Risk: a state of uncertainty where some possibilities involve a loss or catastrophe or undesirable outcome. Measurement of risk is a set of quantified probabilities with quantified losses.

How to Measure Anything: (

What are all the possible decision options?

List out all the possible decision making options. For each of these options, be sure to calculate the things to the right.

How much does it positively impact the dominant goal?

What are all the ways this option can go wrong?

What is the probability of success and failure?

You are assessing and sizing risk and reward here.

What are all the ways I can hedge failure?

Does this change the probability distribution?

When does this option need to be executed?

What are my assumptions, and how certain am I about them?

All options, and their assigned probabilities of success and risk will be influenced by the assumptions we make. How certain am I about these assumptions, and how can I reduce that uncertainty?

How am I being fooled?

Ways you’re being fooled:

  • Cognitive bias
  • Incomplete information
  • Deception — how am I being deceived, and what’s in it for them?

Can I generate extra options by changing parameters?

Try changing the goal and seeing what affect that might have on the options available to you. It’s also a good opportunity to clarify why you’re pursuing the goal.

  • If I achieved X, how would I use it?
  • What features of X are the most important for me?
  • Would a weaker/simpler version of X suffice?

Ask others (peers, people you trust) to help you generate extra options.

Go back to ‘What are all the possible decision options’, reassess.

Weigh up the bet and select the best option

Choose the option with maximum reward for minimum risk. This calculation should be done with most of the focus on the dominant goal (and should integrate any other goals you care about too).

Decide when I need to execute this decision. Move to next card.

Decide who to include in the weighing and selecting process:

  • Should you be inclusive in the process and bring people affected ‘along for the ride’? If so, why?
  • Should you be communicating progress to those that are dependent or affected by the decision? If so, why?
  • Who should you use to help with reducing uncertainty, or testing assumptions?

Execute the decision

  • Start measuring the impact of the decision
  • Write down decision assumptions and thought process
  • Decide when to checkpoint/review measurement, and double check assumptions
  • Communicate to stakeholders, folks impacted by the decision

Checkpoint decision

  • Double check assumptions
  • Measure current progress
  • Assess risk, reassign probabilities

Postmortem decision

After the decision has played out, do the following:

  • Figure out who to include in the postmortem, and then:
  • Did you achieve the outcome desired?
  • How did it impact the goal
  • Did your probability distribution of success vs. risk change over time? Why?
  • Can you improve your ‘metacognition’ of decision making?
  • Write this stuff down.

Test Cases

[Removed for example]