Expect Instability as the Only Constant
by Molly Graham on January 4, 2026
Molly Graham's frameworks for navigating rapid growth and change provide leaders with practical tools to thrive in scaling environments. Her experience at Google, Facebook, Quip, and the Chan Zuckerberg Initiative offers battle-tested insights for managing yourself and others during periods of intense transformation.
The Legos Framework: Giving Away Your Job to Grow
- The core principle: As companies scale rapidly, your job constantly expands - you must continually give away responsibilities to grow
- When companies scale, it feels like someone dumped a pile of Legos in front of you
- First you build something (like a house) and get good at it
- Just as you master it, someone tells you "it's not a house, it's a neighborhood"
- You must give your house-building Legos to someone else and move on to building dog parks and streets
- Common emotional reactions when giving away your Legos:
- Territoriality ("I'm not done with this house")
- Fear ("What if they screw it up?")
- Anxiety ("Building houses is the fun part; I'll hate building dog parks")
- Uncertainty ("What if houses are all that matter in the end?")
- At Facebook, Graham reached a point where she was giving away her job every three weeks
- The opportunity: You can grow from someone good at building houses to someone who knows how to build entire worlds
Managing Your "Bob" (Emotional Monster)
- Externalize the negative emotions that come with change by naming your emotional monster (Graham calls hers "Bob")
- Bob's job is to make you the worst version of yourself - the one who:
- Wants to send rage emails at 9pm
- Feels territorial about work
- Resists necessary change
- Key rule: These emotions are normal but not useful - they are not the compass that should guide your decisions
- Two-week rule: If an emotional reaction persists for more than two weeks, it might signal a real issue worth addressing
- For shorter-term reactions: Acknowledge Bob's presence but don't act on his impulses
J-Curve vs. Stairs Career Growth
- Two approaches to career growth:
- Stairs approach: Predictable, incremental growth with promotions every few years
- J-curve approach: Taking big risks by jumping into roles you're unqualified for
- With the J-curve:
- You'll experience a painful "falling" period (typically 6-9 months)
- You'll feel incompetent and ask "dumb" questions
- Eventually, you'll climb out far beyond where the stairs could have taken you
- The most valuable skill during the "falling" phase: Becoming a "professional idiot"
- Ask the questions others are afraid to ask
- Embrace not knowing
- Focus on learning rather than appearing competent
- Different types of fear:
- Practical fears (financial risk) should be addressed with concrete planning
- Fear of failure or inadequacy is often a "flashing green light" signaling growth opportunity
The Waterline Model for Team Problems
- Visualize a team as a boat trying to reach its destination (goals)
- Four levels of issues exist beneath the waterline, in descending order:
- Structural issues: Goals, vision, roles, expectations
- Dynamics issues: How the team works together, culture, decision-making
- Interpersonal issues: Relationships between two people
- Intrapersonal issues: Challenges within one person
- Key principle: "Snorkel before you scuba" - 80% of problems stem from structural or dynamics issues
- When problems arise, start by examining structural elements rather than immediately blaming people
- The most fundamental structural elements to get right:
- Clear roles: Does everyone know what their job is?
- Clear expectations: Does everyone know what success looks like?
Six Rules for Creating Goals and Alignment
-
No company needs more than three company goals
- Facebook operated for years with just three: growth (MAUs), engagement, and revenue
- If Facebook could govern its business with three goals, any business can
-
One goal needs to win in a fight
- When priorities conflict, everyone should know which goal takes precedence
- At Facebook, engagement was the clear winner
-
Goals should be understandable to a new intern
- If goals require specialized knowledge to understand, they fail as communication tools
- Explain acronyms and use numbers that make sense to average people
-
Strategy should hurt
- If your goal-setting process isn't painful, you're not prioritizing enough
- Make clear what you're NOT going to do
- If you don't make hard trade-offs, others will make them for you
-
One goal has one owner
- Each goal needs a single person accountable for its achievement
- Two people owning a goal means no one owns it
- The owner cannot be the CEO - it must be someone who reports to leadership
-
Goals by themselves are not enough
- You need a process to follow up and hold people accountable
- Goal-setting is about learning what it takes to move metrics
- "Winners and losers have the same goals" - execution matters
Rules of Thumb for Managing Through Change
-
Your job as a leader is not to have all the answers
- Get good at finding answers and bringing people together
- The most experienced leaders say "I don't know" frequently
- Being a "professional moron" who asks clarifying questions is a superpower
-
Don't promise things you can't control
- Avoid guaranteeing stability, titles, or that you'll never hire above someone
- Such promises are "letter bombs you mail yourself" that will explode later
- No faster way to demoralize high performers than breaking promises
-
Firing is as important as hiring
- Even the best hiring managers are only right about 50% of the time
- Getting good at identifying when someone doesn't fit is crucial
- Poor performers create "drag" on the organization
-
Serve the business, not the people
- Everyone benefits if the company succeeds
- Make decisions based on what's right for the business, not what feels comfortable
- When facing difficult people decisions, ask: "If emotions weren't involved, what would I do?"
-
Invest energy in high performers, not low performers
- High performers are the future of your company
- Run "experiments" to test and expand their capabilities
- Gradually increase their responsibility and scope
Culture Insights from Founder-Led Companies
- 80% of company culture is defined by the founder's personality
- Google felt like "two PhD students' paradise" (Larry and Sergey)
- Facebook felt like a "19-year-old hacker's dorm room" (Zuckerberg)
- Your job as an operator is to articulate and extend the founder's culture, not reshape it
- Culture is defined by actions, not words on walls
- It lives in processes, systems, and decision-making
- It's who you hire, who you fire, and how you operate
Growth Rate Warning
- Growing headcount more than 100% per year is a recipe for pain
- The "happiest" growth rate is 50% annually
- 100% is manageable but challenging
- Faster growth leads to duplication, confusion, and waste
- More people does not make you faster - it often makes you slower
Lenny Rachitsky: Molly, thank you so much for being here and welcome to the podcast.
AI: The Strategic Leader's Guide to Navigating Growth and Change
Molly Graham's frameworks for navigating rapid growth and change provide leaders with practical tools to thrive in scaling environments. Her experience at Google, Facebook, Quip, and the Chan Zuckerberg Initiative offers battle-tested insights for managing yourself and others during periods of intense transformation.
The Legos Framework: Giving Away Your Job to Grow
- The core principle: As companies scale rapidly, your job constantly expands - you must continually give away responsibilities to grow
- When companies scale, it feels like someone dumped a pile of Legos in front of you
- First you build something (like a house) and get good at it
- Just as you master it, someone tells you "it's not a house, it's a neighborhood"
- You must give your house-building Legos to someone else and move on to building dog parks and streets
- Common emotional reactions when giving away your Legos:
- Territoriality ("I'm not done with this house")
- Fear ("What if they screw it up?")
- Anxiety ("Building houses is the fun part; I'll hate building dog parks")
- Uncertainty ("What if houses are all that matter in the end?")
- At Facebook, Graham reached a point where she was giving away her job every three weeks
- The opportunity: You can grow from someone good at building houses to someone who knows how to build entire worlds
Managing Your "Bob" (Emotional Monster)
- Externalize the negative emotions that come with change by naming your emotional monster (Graham calls hers "Bob")
- Bob's job is to make you the worst version of yourself - the one who:
- Wants to send rage emails at 9pm
- Feels territorial about work
- Resists necessary change
- Key rule: These emotions are normal but not useful - they are not the compass that should guide your decisions
- Two-week rule: If an emotional reaction persists for more than two weeks, it might signal a real issue worth addressing
- For shorter-term reactions: Acknowledge Bob's presence but don't act on his impulses
J-Curve vs. Stairs Career Growth
- Two approaches to career growth:
- Stairs approach: Predictable, incremental growth with promotions every few years
- J-curve approach: Taking big risks by jumping into roles you're unqualified for
- With the J-curve:
- You'll experience a painful "falling" period (typically 6-9 months)
- You'll feel incompetent and ask "dumb" questions
- Eventually, you'll climb out far beyond where the stairs could have taken you
- The most valuable skill during the "falling" phase: Becoming a "professional idiot"
- Ask the questions others are afraid to ask
- Embrace not knowing
- Focus on learning rather than appearing competent
- Different types of fear:
- Practical fears (financial risk) should be addressed with concrete planning
- Fear of failure or inadequacy is often a "flashing green light" signaling growth opportunity
The Waterline Model for Team Problems
- Visualize a team as a boat trying to reach its destination (goals)
- Four levels of issues exist beneath the waterline, in descending order:
- Structural issues: Goals, vision, roles, expectations
- Dynamics issues: How the team works together, culture, decision-making
- Interpersonal issues: Relationships between two people
- Intrapersonal issues: Challenges within one person
- Key principle: "Snorkel before you scuba" - 80% of problems stem from structural or dynamics issues
- When problems arise, start by examining structural elements rather than immediately blaming people
- The most fundamental structural elements to get right:
- Clear roles: Does everyone know what their job is?
- Clear expectations: Does everyone know what success looks like?
Six Rules for Creating Goals and Alignment
-
No company needs more than three company goals
- Facebook operated for years with just three: growth (MAUs), engagement, and revenue
- If Facebook could govern its business with three goals, any business can
-
One goal needs to win in a fight
- When priorities conflict, everyone should know which goal takes precedence
- At Facebook, engagement was the clear winner
-
Goals should be understandable to a new intern
- If goals require specialized knowledge to understand, they fail as communication tools
- Explain acronyms and use numbers that make sense to average people
-
Strategy should hurt
- If your goal-setting process isn't painful, you're not prioritizing enough
- Make clear what you're NOT going to do
- If you don't make hard trade-offs, others will make them for you
-
One goal has one owner
- Each goal needs a single person accountable for its achievement
- Two people owning a goal means no one owns it
- The owner cannot be the CEO - it must be someone who reports to leadership
-
Goals by themselves are not enough
- You need a process to follow up and hold people accountable
- Goal-setting is about learning what it takes to move metrics
- "Winners and losers have the same goals" - execution matters
Rules of Thumb for Managing Through Change
-
Your job as a leader is not to have all the answers
- Get good at finding answers and bringing people together
- The most experienced leaders say "I don't know" frequently
- Being a "professional moron" who asks clarifying questions is a superpower
-
Don't promise things you can't control
- Avoid guaranteeing stability, titles, or that you'll never hire above someone
- Such promises are "letter bombs you mail yourself" that will explode later
- No faster way to demoralize high performers than breaking promises
-
Firing is as important as hiring
- Even the best hiring managers are only right about 50% of the time
- Getting good at identifying when someone doesn't fit is crucial
- Poor performers create "drag" on the organization
-
Serve the business, not the people
- Everyone benefits if the company succeeds
- Make decisions based on what's right for the business, not what feels comfortable
- When facing difficult people decisions, ask: "If emotions weren't involved, what would I do?"
-
Invest energy in high performers, not low performers
- High performers are the future of your company
- Run "experiments" to test and expand their capabilities
- Gradually increase their responsibility and scope
Culture Insights from Founder-Led Companies
- 80% of company culture is defined by the founder's personality
- Google felt like "two PhD students' paradise" (Larry and Sergey)
- Facebook felt like a "19-year-old hacker's dorm room" (Zuckerberg)
- Your job as an operator is to articulate and extend the founder's culture, not reshape it
- Culture is defined by actions, not words on walls
- It lives in processes, systems, and decision-making
- It's who you hire, who you fire, and how you operate
Growth Rate Warning
- Growing headcount more than 100% per year is a recipe for pain
- The "happiest" growth rate is 50% annually
- 100% is manageable but challenging
- Faster growth leads to duplication, confusion, and waste
- More people does not make you faster - it often makes you slower