Giving Feedback: ONFR

Observation: Not controversial
Feeling: Express your story (and say it's your story)
Need: State what you value, what kind of impact you'd like to see
Request: What concrete action they need to take
Speak not just to the current audience, but also to all the people who will come later.

A lesson from Abraham Lincoln's second inaugural address.
The process is more important than the artifact.
Lynyrd Skynyrd played "Free Bird" for 40 years
When one of your talented people does something dumb, don't blame them.

Instead, ask yourself what context you failed to set.

From Reed Hastings
Human operating system

Everyone wants:

  1. Autonomy
  2. Access
  3. Community
Every night before you go to sleep:

  1. Is everyone on my team being fully utilized?
  2. Are they working on the highest leverage things?
You're allowed to change your mind
The intuitive has more weight than the rational.

Michael Dearing
Respond, don't react
Standup should be: optimize the probability that we hit our goals.

  • Standup should be milestones, so:
    • What did you do since last standup?
    • What will you do before the next standup?
    • Are you blocked?
    • Confirm / revise time estimates
  • AFTER standup: meet, if necessary, to replan
  • My role: Ensure standup happens
  • Your role: do your work
  • Not a management status meeting.
Praise in public, criticize in private
Culture is the way people treat each other.
- Bill Campbell
Intuition

Fast, automatic, low cognitive load, error-prone.

Reason

Careful, slow, high cost.
For interviews:

  • Values (life story)
  • Flight risk
  • Emotional investment
  • 1-10, contingent upon ___
  • Bullet or barrel
  • Superpowers
  • Deadly Risks
Show me > Tell me a story > Description
There is always a better way
Persuasion = credibility (credos) + emotion (pathos) + logic (logos)
CTO: Be the technical headlights of the company
Sports metaphors assume a zero-sum game. Be careful.
The core of managerial leverage, according to Andy Grove:

  1. Information gathering
  2. Nudging
  3. Decision making
  4. Being a role model


  1. Schedule
  2. Quality
  3. Features

Pick two.
Engineers think in terms of implementation models, but users think in terms of conceptual models.
Be highly aligned and loosely coupled.

(Big + fast + flexible)
Speed kills detail and increases noise.

Frank Chimero, The Inferno of Independence
What are the top 3 most difficult things right now?
Orville Wright did not have a pilot's license

From Orbiting the Giant Hairball
Use a Drake equation.

N = Ax * Bx * Cx ...
Do something every day to make tomorrow better.
OKRs are the solution to every management problem, always, in every company.
Conduct the real meetings before the official meeting
  1. System 1: Gut
    1. First principles
    2. Past experience
  2. System 2: Thinking
    1. Stored, accessible analytics
This has all happened before
Engineering knows what's possible, so they are an important input to the product discovery process.
A problem solving style:

  • Understand the problem
  • Manage the flow of ideas
  • Maintain quality


Understand the problem

  • Success or failure often turns on minuscule differences in problem definitions
  • Leaders recognize this, hackers get bored once "it works"
  • Encourage everyone to "read the spec", or understand the problem
  • Resolve arguments by referring back to the original problem
  • Get to a common understanding of the problem
  • Is the argument about a difference in the solutions or a difference in the problem it solves?
  • Ask!

Manage the flow of ideas

  • Contribute a clever idea to the team
  • Encourage copying of useful ideas
  • Elaborate on an idea that a teammate contributed
  • Drop your idea in favor of one the team wants to develop
  • Refuse to drop an idea until everyone on the team understands it
  • Resist time pressure, listen when other people explain ideas
  • Test ideas contributed by other people
  • Withhold quick criticism of ideas -> keeps them flowing
  • Make it clear that you are criticizing the idea, not the person
  • Test your own ideas before offering them
  • Encourage the team to drop ideas that worked before, but no longer
  • Revive a dropped idea later

Maintain quality

  • Measure quality as the project progresses
  • Design tools / processes to test quality as you build a solution
  • Measure the speed of the implementation, compare to schedule, be prepared to change the solution procedure
  • Step back and assess viability, refresh perspective
  • Check ideas with the customer before building
  • Restore morale when an idea collapses


Smooth is fast
Premortems

To debug a plan:

  1. 2 people write a story (not a list) about the current plan, fast-forwarded and assuming disaster
  2. 2 people write a story about how the current plan will lead to success
  3. Retrospective: Should we change anything about the plan?
Meetings are not an activity, they are a medium.

Andy Grove
Everything is either an incremental improvement or a big bet.

Adam Wiggins
Simplify or diversify?
Context > Control
Here’s how to get people to change:

  • Nobody Likes To Be Told What To Do: When we tell people what to do, they resist. Ask questions instead. Support “change talk.” Ignore “sustain talk.” 
  • Use “OARS”: Ask open questions. Affirm their strengths. Lots of reflective listening. Summarize. 
  • Shift The Ratio Of Change Talk To Sustain Talk: Ask questions that provoke DARN: Desire, Ability, Reasons and Need. This gets them focused on changing instead of sustaining. 
  • Develop Discrepancy: Ask questions that evoke the discrepancy between their values and behavior. 
  • Evoke Hope: Get them to reflect on previous successes in changing behavior. Get them to apply those lessons to the current issue. 
  • Planning: Ask “So what’s next?” Use questions to get them to clarify and make a concrete plan. Summarize. Present possible challenges so they can troubleshoot and solidify the agenda.

Nobody Likes to be Told What to Do

As Bern Williams said: “Unsolicited advice is the junk mail of life.” But, more importantly, Goethe said, “If you treat an individual as he is, he will stay as he is, but if you treat him as if he were what he ought to be and could be, he will become what he ought to be and could be.”

DARN

  1. Desire: I want to lose weight
  2. Ability: I want to run a marathon but I don't think I can do it
  3. Reasons: I want to exercise so I have more energy
  4. Need: I need to get in better shape

  • "How do you want your life to be different a year from now?” provokes Desire statements. 
  • “If you did really decide you want to lose weight, how could you do it?” can get Ability responses. 
  • “Why would you want to get more exercise?” gets you Reasons. 
  • “How serious or urgent does this feel to you?” evokes Need.

Develop Discrepancy

Research show one of the primary motivators for change is someone noticing the discrepancy between their goals and their behavior. It’s uncomfortable. And to relieve that discomfort, our behaviors usually change before our values do.

Evoke Hope

The best way to get people’s confidence up is to ask them about previous challenges they successfully overcame.

Planning

At this point you’ll want to lean back, tent your fingers and smirk. Don’t. Instead, be happy for them.

From https://bakadesuyo.com/2022/05/get-people-to-change/
Lack of feedback = affirmative!
A good decision today is better than a great decision tomorrow.

Lynn Reedy
No-Problem Syndrome

Early NPS detection:

  1. Describe very difficult problem
  2. Respondent says "no problem"
  3. You say: "Great! Can you describe my problem you'll solve?"

If respondent describes problem, even erroneously, it's not NPS - just enthusiasm.

If respondent describes proposed solution, it's NPS.
Lead by Example
Recruiting isn't transactional. Build relationships.
Cognitive distortions

Personal Exceptionalism

"I am special"

Dichotomous Thinking

"X is shit. Y is genius."

Correct overgeneralization

"I see a dot and draw a line. And it's the right line."

Blank-canvas thinking

"Paint by numbers isn't art. I want to make art."

Schumpeterianism

"I am a creative destruction machine."
The Cognitive Behavioral Model

Filtered Lenses:

  1. Worldview
  2. Future Expectations
  3. Self-image
General Management Toolkit

  1. McCalllum's Key Questions
  2. OKRs, Black Box, Leverage
  3. Cognitive Distortions
  4. S1 and S2
  5. Drake's Equation
  6. Minto Pyramid Principle
Early on, people have jobs. Later, jobs have people.
Factory or studio?
Perfect is the enemy of good enough.
Cash is life
Choose your battles
Monthly Management Exercise

  1. Pick 2 activities to eliminate from your calendar
  2. Pick 2 activities to add to your calendar
  3. Pick your two weakest McCallum questions
  4. Your OKRs for the next month
Where other companies stop, we find a way.
- Lynn Reedy

Amplify comparative advantages, shock-absorb for deadly risks.
Transitions are almost always signs of growth, but they can bring feelings of loss. To get somewhere new, we may have to leave somewhere else behind.
Set ridiculous goals, then work backwards.
When someone fails:
  1. Set the bar high
  2. Point out the miss
  3. Be supportive

I have really high expectations of you. I like it when you _____, but you didn't do that in this case. Is everything OK?
Are you:

  1. Results oriented?
  2. Internally directed?
  3. Other focused?
  4. Externally open?
Plans are useless, but planning is essential.

Dwight D. Eisenhower
One theory of a great small team

  1. Presence
  2. Density
  3. Intensity

(From Bloc employee peloton analogy)
To develop a long-term relationship with an engineer:

  • What skills do they want to improve?
  • What technical and non-technical experiences do they want to have?
  • How do they want to increase their scope of impact at the company?
3 great obstacles to innovation:

  1. Self blindness
  2. No-problem syndrome
  3. Belief in the central dogma of academic psychology
Tap into universal values
Happy = productive
Transparency is a serious accelerant.
Every argument ever:

  1. Situation
  2. Complication
  3. Question
  4. Answer

(Minto Pyramid Principle)
Org Design

  1. Objectives for the team?
  2. What chain of events should repeat over and over again?
  3. What humans and machines are available?
  4. How to combine them so that:
    1. Cost > Price > Perceived value?
Composting

I failed to _____. I observed that _____. Next time, I might try _____.
A healthy outcome comes from a healthy pipeline.

Michael Dearing
Empathy -> Respect -> Influence -> Colleagueship
McCallum's Questions

  1. How do you get a group of people working together towards common goals?
  2. How do you give the right amount of responsibilities?
  3. How do you make sure the job gets done?
  4. How do you know how things are going?
  5. How do you do all of this with respect for others?
One outstanding employee gets more done and costs less than two adequate employees. Pay well.
Build relationships before you need them.
5 Sacred Duties of a Manager

  1. Communicate priorities and common goals
  2. Divide work and authority
  3. Deliver results
  4. Evaluate performance
  5. Repeat and improve

Give it a minute to sink in before you respond.
Control your destiny, or someone else will.

Jack Welch
Some people "get it", and some people don't.
3 Measures of Leadership

  1. Clarity of thought and communication
  2. Judgement about people
  3. Personal integrity and commitment
Thinking is bottoms-up. Communicating is top-down.
Barbara Minto
Empathy exercise

I'm having trouble _____. I need help with _____ so that I can ______.
Org structure is a prototype. Obsolete the day it's created.
Remember: Your message is still new to many people
Objectives

  1. Must-dos to execute strategy and ambitions
  2. Process based or outcomes based
  3. Nests together at all levels

Key Results

  1. Measurable
  2. Time-bound
  3. Ambitious (60% achievable)
Know what you're looking for before you look for it.