Latest / Safe As Podcast / E82: Why Good Leaders Make Terrible Decisions (The Science of Bias)

Transcript

E82: Why Good Leaders Make Terrible Decisions (The Science of Bias) · Safe As Podcast

• 0:00
  A lot of leaders can now name common decision biases like
• 0:03
  confirmation bias, anchoring, loss aversion.
• 0:06
  But the authors in this article today point out something
• 0:09
  important. Awareness alone hasn't improved
• 0:12
  decision quality. So the practical question is,
• 0:14
  when a team brings you a major recommendation, what can you do
• 0:17
  that actually helps mitigate bias?
• 0:20
  Good day everyone, welcome to Safe as I'm Doctor Ben
• 0:23
  Hutchinson. Today's article is from Karneman
• 0:25
  ET al, 2011 titled Before You Make That Big Decision,
• 0:29
  published in Harvard Business Review.
• 0:31
  The article explains that we think in two modes, System one,
• 0:35
  which is the fast, intuitive one or called Type 1, and System 2,
• 0:39
  which is slow and deliberate, or also called Type 2.
• 0:42
  Type 2 comes online when stakes are high or careful reasoning is
• 0:47
  needed. But most of the time, Type 1 is
• 0:49
  doing the work. It works in the background,
• 0:51
  forming impressions, using learn patterns, and creating a story
• 0:55
  that feels coherent. Type 1 is designed to produce a
• 0:57
  single, coherent interpretation of events, and as it does, it
• 1:01
  tends to suppress alternative stories.
• 1:03
  Intuitive errors don't feel like errors while they're happening.
• 1:07
  We almost never catch ourselves making them.
• 1:09
  That's why just knowing your biases isn't a reliable fix.
• 1:13
  So the authors propose adding quality control to big
• 1:16
  decisions. They propose a 12 question
• 1:18
  checklist that reviews the process behind any
• 1:21
  recommendation, not just the content.
• 1:23
  So you're better aren't to detect likely distortions in the
• 1:26
  team's reasoning before you approve the call.
• 1:28
  So on to the 12 questions. On the first pass, ask yourself
• 1:33
  one motivated errors. Is there a reason to suspect
• 1:36
  errors driven by the team's self interest?
• 1:39
  This isn't mainly about deliberate lying.
• 1:41
  Self deception and rationalisation are far more
• 1:44
  common if the team has unusually strong incentives, money, power,
• 1:48
  reputation, career options, apply extra scrutiny to the
• 1:53
  effectoristic Have they fallen in love with the proposal?
• 1:56
  When people like an idea, they tend to minimise the risk or the
• 1:59
  cost and exaggerate the benefits.
• 2:01
  If yes, tighten the rest of the review. 3 Groupthink.
• 2:05
  Were there dissenting opinions and were they explored?
• 2:09
  No Dissent on a complex decision is a warning sign.
• 2:12
  Discreetly solicit dissenting views if needed.
• 2:15
  Start with the person with the least experience or the quietest
• 2:18
  voice. Second pass.
• 2:19
  Ask the recommenders some questions to see how they
• 2:21
  respond. So 4 Saliency bias Is the
• 2:25
  diagnosis overly shaped by a memorable success analogy?
• 2:28
  Ask for multiple comparable examples and analyse how similar
• 2:32
  they really are. 5 Confirmation bias.
• 2:35
  Were credible alternatives genuinely considered?
• 2:38
  Ask what options were evaluated when they were rejected, and
• 2:42
  whether the team actively looked for disconfirming evidence, not
• 2:45
  just confirming evidence. 6 Missing info.
• 2:48
  If you had to make this decision again in a year, what
• 2:51
  information would you want and can you get more now?
• 2:54
  7 anchoring Do you know where the numbers come from first
• 2:58
  numbers on the table and whether estimates are anchored to
• 3:01
  guesses or straight line history?
• 3:03
  If it matters require re anchoring using independent
• 3:06
  models or benchmarks and requests new analysis. 8 The
• 3:10
  Halo effect is the team assuming that because a person or a
• 3:14
  company or some sort of approach succeeds in one area, it will
• 3:17
  succeed in another. Test comparability and ask for
• 3:19
  counter examples include less successful cases 9.
• 3:23
  The sunk cost fallacy. Are they overly attached to past
• 3:26
  decisions? Use the new CEO lens.
• 3:29
  If you hadn't made the earlier call, would you invest them now
• 3:32
  and pass? 3.
• 3:34
  Evaluate the proposal, or in other words, test the realism of
• 3:37
  it 10. Base case optimism.
• 3:40
  Is the base case overly optimistic?
• 3:42
  Take an outside view using comparisons with similar
• 3:45
  projects rather than relying on only an insider view.
• 3:49
  Perspective 11 Disaster neglect Is the worst case Bad enough?
• 3:54
  Run a pre mortem, assume the worst has actually happened and
• 3:57
  build the story of how it happened to surfers.
• 4:00
  Risks you may not have experienced yet. 12 Loss
• 4:03
  aversion is the team overly cautious, avoiding
• 4:05
  responsibility for a potential failure.
• 4:07
  Leaders may need to share responsibility for risk or
• 4:10
  realign incentives so reasonable risk taking is possible.
• 4:13
  The core message from their 12 point checklist is about
• 4:16
  improving decisions, is about disciplined process and not
• 4:20
  necessarily about individual brilliance.
• 4:22
  That's it on safe as. Thanks everyone.
• 4:25
  Don't forget to like, subscribe and comment on my YouTube
• 4:28
  channel and links in show notes if you'd like to support the
• 4:31
  channel.