Cognitive Biases in Engineering Organizations (Etsy)

Cognitive Biases in Engineering Organizations
---------------------------------------------

- Johnathan Klein, Etsy

                                ,-.
                                |  |
                             ,--': `--.
                             `--,  ,--'
                                |  |
                                |  |
                                |  |
                                ;--|
                                ;  :
                               /    \
                            ,-v|::::|    ,`^,
                        v.   )/:::::|_  `)v(
                     ,-'),\  |:::::/ ^\   ||
                      )(||   |:::::"\/"  | |                    _
                     | |||   )`----'||   | |                 __( )
                  __ | || \  |._____||   | |   __           (  )(
                 (  )| || | | v v v|'`|  | |  (  )           )(| |
                  )( | |I |(^)v v v|,.| .' |   )(            | | |
              __  ||/  |I | )|v v v|I | |  |  | |           _| | |
             (  ) ||mmm|   \||v v v|I '||  |  | |          (  )| |
              ||  ||/v,+--'|||v(^)v|I  || ^|  | |           || | |
              ||  | \ |^ ^^`'`vv||v|vvv|| ^'.|  |           /| | |
              | \ |-| |^ ^^/  \v||/^ ^^|/\^^||  |          .'| | |
              |  || | |    |  |v|||^ ^^|  \^|   |          | '.|.'
              |  \+-\ |^ ^^|__| |||    |   \|   |          |  | |
              |--|   \|^ ^,+  |+--+^ ^ |    Y _ | _        |  | |
              |  |   \|,-.:|  ||  |^ ^ |    |(.) (.)  _   |   | |
              |   \  \|; `;|  \/--+    | | /--.-._|_ (.)  |   | |
              |   |   \/ ( |   |  |^ ^d| |/| /|   | `-,-- |   | |,
              |   |   \ \ \|   |  | ^|8||o |o |  /|  /|   |   | '.
   -hrr-      |   |  _,\--.|   |   | `"||| || | o | o |  |    '. |
              |\,-|-' _|--.|   |   |^ ^| | || | | | | | o|    ||  |
             /| |,|,-'   \ |._ |   |^^^|/O\/^\|/^\|/^\|/ |    ||  .
       _.Y._/ |   |       \|  `|   |  ^|_ / O Y O | O\/ |   ,'||  |
     ,'     `\|   |    |   Y   |`--|  |   |---|.__|  /  |_,'  |   |
  ,-,         )   |   -+-  |   |   | /^\  |   |   |  /  |    ,.   |
 (  \       /\|       ,|.  |   |   |/   \/^\ /\  /\ /\\ |   | _'  |
  \  \ \\/\//\    _,-_/  `.|   |    | o | o Y  \/  \//\|    ' [  /|
   ) /\,-       ,:--'::-.  `.  |    |,-.|__ | o| o /\\\|   |   |/||
  (  \   \/\/\,( ...:::::`-. `.     ( . |. ),-.|,- \\//\   |  //||
   `--`-.___  /\-,----,._   `-.`.   /\  `-' `._| .//\\\/ _|.,'/||
            `--//// //|  `-._  `-\  | \_    |   _/\\\///\    /|||
              //// //|,: \\ \\--._\ |_| `---|__/  //\\/\\   / ||
             `-'`'//_|,   \\ \\ \\ /v\/\,      `.//_|| | | /|
                 (_)       \\ \\ \\-._   \/ \;.,.    `-^',' |
                           (_)) \(_)  `---.__  ` \/\/\  (
                              `-'            `---._  /   \
                                                   `(     )
                                                     `---'


- ref. Thinking Fast and Slow

Engage system 2 thinking
10% videos on iOS Youtube

Projection Biases
- sales thinks that selling is top priority etc
    Solution
        - Empathy
        - Have people switch roles

Planning Fallacy
    - people always pick the best case version of the role
    Solution:
        - Prototyping
        - Use information from similar ventures
        - Planning poker
            - See slides

        Fibonnaci gives a descrete set of numbers that get less specific as they go up - estimates above a certain threshold are represented as highly speculative

Bandwagon effect:
    - Don't prime the estimate

Sunk Cost Fallacy

Hyperbolic Discounting
    - The Stanford Marshmallow Experiment
        - https://www.youtube.com/watch?v=x3S0xS2hdi4
        - Children who are able to wait longer generally have better outcomes\

Fundamental Attribution Error
    - Blameless postmortems
    - 'When someone cuts you up they're a jerk, when you cut them up, it's because you're in a hurry'
    - Preys on Loss Aversion
    -

Denominator neglect


Anchoring effect
 - highlight a less expensive plan by placing it next to a much more expensive one

Freemium model
    - In game purchases
    - Sum cost fallacy
    - Same reason you'll sit through a movie you hate but you paid for

Bayesian Reasoning
    - anchor judgement on plausible base rate
    - question the importance and relevancy of your evidence
    - make your expectations line up with reality


Predictably Irrational
You are not so smart
Harry Potter and the Methods of Rationality



Doing a support rotation - really useful for seeing what Etsy as a business is like on the ground

Knowing the biases helps - helps question responses

See: Less Wrong community