VIDEO - “I like being scared”- Molly Graham’s frameworks for rapid career growth - Molly Graham

My Notes

• substack: Lessons | Molly Graham | Substack
• 80% of the issues on the team are caused by structural or dynamic aspects.
  • This means one should not start from the bottom, looking for personal or interpersonal issues.
  • Instead, the focus should be on what can be changed in the structure → the setting of expectations, or the team dynamics
• What to ask and then look for the answer when your team is not performing?
  • How do you organize this group of people so they know in which direction their boat is going?
  • then go and talk to everyone in the team and ask: does this person actually know what is expecting to them? If not, go back and set the clear expectations
  • Specifically ask the question: What do you think your job is? What do you think you have been hired for?
• what is the point of the goals? → communication tool designed to bring clarity into the company
• sign of good strategy: good strategy should be painful → if you are not making decisions that are painful, you are not helping people prioritize their time
• People will naturally show up every day and do their work → Either you are very clear with them about what the priorities are, or they will prioritize the things for you
• never promise anything out of your control → eg: cannot promise you will be my manager forever → that may change, cannot promise onboarding will be smooth, depends on person etc

Mindmap

Main Idea of the Text

Molly Graham shares battle-tested frameworks for leadership during rapid scale, drawing from her time at Google, Facebook, and Quip. Key concepts include "Giving Away Your Legos" (delegating to scale), the "Waterline Model" (diagnosing team issues via structure vs. people), and specific rules for high-impact goal setting and managing high performers.

Relevance to You

This is 5/5 (Maximally Relevant). As a newly promoted Senior EM driving a shift from output-driven to outcome-driven development, Graham’s frameworks on role clarity ("Waterline") and scaling yourself ("Legos") are critical tools for optimizing your R&D department and growing your direct reports without increasing headcount.

Key Learnings for Obsidian

• Give Away Your Legos: To grow, you must hand over tasks you are good at to others; if you hold on, you become the bottleneck.
• The Waterline Model: 80% of team problems are structural (goals/roles), not personal. "Snorkel" (fix structure) before you "scuba" (fix people psychology).
• Manage "Bob": Bob is the voice of insecurity/ego during change. Wait 2 weeks before acting on strong emotions; if they persist, it's real.
• J-Curve Careers: High-growth careers look like jumping off a cliff (high risk/failure) rather than walking up stairs.
• Goal Limit: No company/team needs more than 3 goals. One goal must "win in a fight" (be the clear priority).
• Strategy Should Hurt: If prioritizing goals isn't painful, you haven't prioritized enough.
• Ownership: One goal, one owner. Two owners mean zero owners.
• Escalation is a Tool: Escalation isn't tattling; it's resolving a deadlock when two peers lack the authority/context to decide.
• Serve the Business: Make decisions based on "what would I do if there were no emotions involved?" then execute with kindness.
• Focus on High Performers: Don't ignore top talent; run experiments to stretch them, as they generate the highest ROI.

Main Ideas

• Snorkel Before You Scuba (Waterline Model)

  "80% of problems on teams actually happen because of structural issues or dynamics issues... Your only goal as a manager if you do nothing else is clear roles and clear expectations."
  Why it matters: Before assuming a developer is underperforming or a "bad fit," you must rigorously audit if their role and success metrics (outcomes) are explicitly defined.

• Giving Away Your Legos

  "You have to grow as fast as your company is growing if you really want to take advantage. Both learning to give away what you've gotten good at and move on to the next shiny pile of Legos."
  Why it matters: As you moved from EM to Senior EM, you must stop building the "houses" (tasks you mastered) and start building the "neighborhoods" (R&D strategy), or you will stunt your team's growth.

• Strategy Must Be Painful

  "Strategy should hurt... if your goal setting process is not painful, then you're not prioritizing heavily enough."
  Why it matters: To move from output (shipping features) to outcome (delivering value), you must aggressively cut "good" ideas to focus only on the "best" ideas that drive metrics.

• One Goal, One Owner

  "One goal has one owner. You have a number. That number has a name next to it. If you cannot do that work, you haven't done the most important work."
  Why it matters: Shared ownership dilutes accountability; assigning a specific engineer to a specific key result is the fastest way to drive autonomy and product-mindedness.

• Invest in High Performers

  "High performers are actually the future of your company... I run experiments... I'm going to see if they can do this without, you know, with less guidance or support from me."
  Why it matters: Instead of spending 80% of your time fixing low performers, you should focus on stretching your best devs to become the "full-stack product engineers" your vision requires.

Actionable Ideas

• Audit Role Clarity Immediately

  "When I take over a team... part of what I'm asking is what do you think your job is? What number were you hired to drive? Because what you'll find is often like their picture is different than your picture."
  Action: Ask your 10 DRs to write down exactly what they think their job is and what metric defines their success. Compare it with your definition to find the "structural" gaps.

• Separate "Bob" from Reality

  "My rule of thumb... is give it two weeks... anything that lasts longer than two weeks is actually something you should pay attention to."
  Action: When you feel a flash of anger or insecurity regarding a decision by the CTO or a conflict in the team, wait 14 days before sending the email or making a structural change.

• Reframe Escalation for Your Team

  "What escalation is is we disagree. Neither one of us has enough power to make this decision. Let's go to someone who does... As soon as you are stuck, escalate."
  Action: Explicitly tell your teams that escalating to you is a requirement if they cannot agree on a technical approach within 24 hours, to maintain velocity.

• Make the "No Emotion" Decision

  "If there were no emotions involved if this person had no negative reaction to this what would I do... and then that's the thing you should do."
  Action: Use this heuristic when deciding whether to keep a struggling engineer or cancel a project that team members love but isn't driving outcomes.

• Define the "Winner" Goal

  "One goal needs to win in a fight... If I'm sitting down and asking how do I prioritize my time on a given day, I need to know what is the most important thing."
  Action: Look at your R&D OKRs. Identify which specific metric (e.g., User Retention vs. New Features) wins if a dev has to choose between two tickets today.

Step-by-Step Plan for Senior EM Context

Phase 1: Diagnosis & Structure (The 80/20 - Do this immediately)

1. The "Elephant" Exercise: In your next 1:1s with your 10 Direct Reports, ask: "What do you believe is the single most important number/outcome you are responsible for?" compare this to your expectation. If they answer with a task ("I code React components"), you have a structural problem to fix to move to outcome-driven.
2. Audit the "Legos": List the top 3 activities you do weekly. Identify which one you are doing just because you are good at it (e.g., code reviews for a specific feature, Jira grooming). Hand this over to a senior engineer immediately to create space for your new strategic responsibilities.

Phase 2: Operational Hygiene (Weeks 2-4)
3. Refine OKR Ownership: Review your SMART OKRs. Ensure every Key Result has one specific human name attached to it, not a team name. This forces the "product engineer" mindset.
4. Institutionalize Escalation: Announce a new principle: "If two engineers or an engineer and PM disagree for more than 1 day, they must come to me together to unblock." Frame this as efficiency, not failure.

Phase 3: Growth & Optimization (Ongoing)
5. High-Performer Experiments: Select your top 2 engineers who have "product engineer" potential. Design a "cliff" for them—give them a vague problem (e.g., "Increase efficiency of the onboarding flow") with no tickets, just the goal, and let them struggle/solve it to build their autonomy.
6. The 2-Week Rule: Implement a personal protocol using Obsidian. When you feel a strong negative emotion regarding a strategic shift, write it down in a daily note. Tag it #review-in-2-weeks. Only act if the tag resurfaces and the feeling persists.

This article was originally published on https://craftengineer.com/. It was written by a human and polished using grammar tools for clarity.
--

Follow me on X (Formally, Twitter). Or read my stories on engineering management, and how to be a better engineering leader on Vibe Manager Blog.