Work on What Matters
selt time-management prioritization leverage
Status: Notes complete
Overview
Staff engineers face a deceptively hard problem: their time is scarce, their skills are in demand everywhere, and almost every request can be dressed up to sound important. This section addresses how to distinguish genuinely high-impact work from the many traps that feel productive but consume finite resources — time, energy, attention, and social capital — without creating proportional value.
The core challenge is not busyness. Most Staff engineers are extremely busy. The challenge is ensuring that busyness translates into organizational impact. Larson frames this as a discipline of subtraction: ruthlessly identifying and eliminating low-leverage activities so that high-leverage work can actually happen.
Core Concepts
Snacking — Easy, low-effort work that generates a feeling of productivity without meaningful impact. Examples include attending meetings where your presence adds nothing, reviewing low-stakes PRs out of habit, or polishing documentation that nobody reads. Snacking is dangerous precisely because it is satisfying and socially acceptable.
Preening — Highly visible work that signals effort and ownership but does not move the organization forward. Rewriting a service in a trendier language without a compelling technical case, or volunteering for high-profile committees without delivering substance, are classic examples. Preening differs from snacking in that it actively consumes organizational credibility; people eventually notice the ratio of visibility to results.
Chasing Ghosts — Pursuing technically interesting work that nobody in the organization actually needs. This trap is especially common for technically excellent engineers who have outgrown an entry-level appetite for novelty. A ghost project can consume months before anyone asks whether the output will be used.
Existential Risks — Problems that, if left unresolved, could cause serious harm to the organization: a security posture that is one breach away from regulatory action, a scaling bottleneck that will fail under the next product launch, or a team structure that is burning out its best people. Staff engineers are often the only ones with enough cross-organizational context to see these risks clearly. Failing to address them is a failure of role.
Work Only You Can Do — The highest-leverage category. At any given moment there will be work that genuinely requires your specific combination of context, relationships, institutional memory, and technical depth. No one else can substitute. This work deserves priority above everything else.
Finite Resources — Time, energy, attention, credibility, and social capital are all finite. Spending any of them on snacking or preening is not just neutral — it crowds out high-leverage work. The resource most commonly underestimated is attention: deep engagement on a hard problem requires cognitive reserves that are depleted by context-switching and low-stakes commitments.
The Four Categories of Work
Larson’s taxonomy maps work roughly onto two axes: organizational impact (does this matter?) and personal leverage (am I the right person?).
| Category | Impact | You as the right person? | Verdict |
|---|---|---|---|
| Existential risks | Very high | Often yes | Do it — urgently |
| Work only you can do | High | Yes by definition | Do it |
| High-impact, others can do | High | No | Enable/delegate |
| Snacking / Preening / Ghosts | Low | Irrelevant | Decline or minimize |
The discipline is not just doing more of the top rows — it is actively saying no to the bottom row, which requires both self-awareness and organizational courage.
Resisting Snacking
Snacking is reinforced by several organizational dynamics:
- Completion bias: finishing a small task feels better than making incremental progress on a large ambiguous one.
- Social pressure: declining visible requests (meetings, reviews, committees) can feel rude or disengaged.
- Uncertainty avoidance: snacks are predictable; high-leverage work often involves ambiguous, uncomfortable problems.
Larson’s advice is to notice the pattern — not to feel guilty about individual snacks — and rebalance your time allocation deliberately. A useful diagnostic: could a competent engineer two levels below you do this work just as well? If yes, it is probably snacking.
Preening and Credibility Erosion
Preening is more insidious than snacking because it consumes social capital. When peers and leadership notice that your high-visibility contributions do not produce proportional outcomes, your ability to advocate for important work erodes. Staff engineers rely heavily on informal influence; credibility is the currency that makes that influence possible.
Signs you may be preening:
- You gravitate toward projects where your contribution will be publicly recognized over projects where it will be quietly essential.
- You find yourself rewriting things that work adequately rather than solving problems with no existing solution.
- You volunteer for steering committees but rarely complete the follow-through.
Finding Existential Risks
Identifying existential risks requires the kind of cross-organizational visibility that Staff engineers are positioned to have but must actively cultivate. It requires:
- Talking to enough people across teams, product, ops, and leadership to see systemic patterns that no single team sees.
- Reading signals that others ignore: repeated incidents that get patched rather than fixed, hiring that cannot keep pace with attrition, technical debt that has begun to block feature delivery.
- Having the courage to name uncomfortable problems before they become crises, including problems that implicate decisions made by senior leaders.
Addressing existential risks is often unglamorous. The work may involve months of coordination, writing documents no one initially wants to read, and pushing on problems that make people defensive. It is also precisely the kind of work that justifies the Staff engineer title.
Writing vs. Talking
One of Larson’s most concrete recommendations is to write more and talk less. Written artifacts — design docs, strategy documents, postmortems, ADRs — scale. A good document can influence dozens of engineers and decisions long after the meeting that would have covered the same ground is forgotten.
Writing also forces precision. It is easy to maintain a vague impression of expertise in conversation. Writing exposes the gaps. Staff engineers who invest in writing compound their influence over time; those who prefer talking tend to become bottlenecks because every decision requires their live participation.
Avoiding the Bottleneck Trap
A Staff engineer who is “helpful” in the sense of inserting themselves into every decision creates a different kind of organizational damage: they become a single point of failure. Teams wait for their input, decisions slow down, and the engineer’s own high-leverage work gets crowded out by coordination overhead.
The remedy is to invest in systems rather than approvals:
- Write the design principles so teams can make good decisions without asking you.
- Mentor the engineer who will own the domain so they develop judgment, not just instructions.
- Build review processes that surface the decisions requiring your attention, rather than routing all decisions to you.
Energy and Attention Management
Larson treats attention as a first-class resource alongside time. Key principles:
- Protect deep work blocks. Complex cross-organizational problems cannot be solved in the gaps between meetings. A Staff engineer who has no uninterrupted time is effectively unable to do Staff-level work.
- Match task type to energy level. Shallow coordination work (reviewing documents, answering Slack) can be batched during low-energy periods. Reserve high-energy periods for the ambiguous, hard problems.
- Recharge is work. Sustainable high-leverage output requires deliberate recovery. Burning through energy reserves leads to the kind of diminished judgment that causes Staff engineers to default back to snacking.
The Discipline of Saying No
Saying no at Staff level is not a personality preference — it is a professional responsibility. Every yes to low-leverage work is an implicit no to high-leverage work. Useful heuristics:
- “What will I not do if I do this?” — make the trade-off explicit before committing.
- Offer an alternative route: rather than just declining, point the requester toward the person or resource that can actually help.
- Negotiate scope, not just effort: “I can’t own this, but I can give you 30 minutes to get you unblocked” is often the right answer.
Key Takeaways
- Time, energy, attention, and social capital are all finite — every allocation is also a de-allocation elsewhere.
- Snacking (easy, low-impact) and preening (visible, low-impact) are the two dominant failure modes of Staff engineer time management.
- Chasing ghosts — interesting work that serves no organizational need — is a third trap, especially for technically excellent engineers.
- Existential risks deserve urgent priority; Staff engineers often have unique visibility to identify them and unique leverage to address them.
- Work only you can do is your highest-leverage use of time; protect it fiercely.
- Write more, talk less: written artifacts scale your influence without requiring your ongoing presence.
- Avoid becoming a bottleneck: build systems and people, not approval dependencies.
- Saying no is a professional responsibility, not a personality trait — make trade-offs explicit and offer alternative routes.
- Protect deep work: uninterrupted time is a prerequisite for Staff-level thinking, not a luxury.
- Unglamorous, high-impact work is frequently the most valuable work a Staff engineer can do — resist the pull toward visible-but-low-impact contributions.
Related Resources
- sec01-staff-engineer-archetypes — understanding your archetype shapes which high-leverage work fits you
- sec03-writing-engineering-strategy — written strategy is the scaled output of doing the right work
- sec04-managing-technical-quality — managing quality is a concrete example of high-leverage, often unglamorous work
- ch03-leading-big-projects — complementary view on prioritization at Staff level
- ch05-being-visible — how to remain visible without preening
Last Updated: 2026-05-30