Focus

The most important and most difficult founder discipline. Altman distills the best founders to two words: “focus and intensity.” Startups don’t die from doing too little — they die from doing too many things at once, none of them well enough.

Why Focus Is So Hard

Startups face infinite demands and finite resources:

  • Customers want 10 features; you can build 2
  • Three growth channels look promising; you can invest in 1
  • Investors want metrics, product wants polish, hiring needs attention
  • Everything feels urgent; very little is actually important

The natural response is to spread across everything. This is the wrong response. Spreading thin means nothing gets done well enough to matter.

The Frameworks

Altman: Focus and Intensity

“If distilled to only two words, ‘focus and intensity’ really apply to the best founders.”

  • Say no frequently — don’t attempt multiple major initiatives simultaneously
  • “I have never, not once, seen a slow-moving founder be really successful”
  • Evaluate every activity: “Is this optimizing growth?”
  • Early founders with publicists almost always fail — focus on the company, not yourself

Rabois: Editing, Not Writing

Your job as leader is to edit, not write — simplify the organization down to 1-3 core priorities.

  • Eliminate unnecessary meetings, processes, and features
  • If you have more than 3 priorities, you have none
  • The more you simplify, the better the organization runs

Seibel: The 90/10 Rule

From seibel-yc-essential-advice: seek 90% of the objective with 10% of the effort.

  • A timely adequate solution beats a perfect delayed one
  • Find the version of the task that takes 10% of the time and delivers 90% of the value
  • Apply ruthlessly to every decision: is this the 10% version or the 100% version?

Graham: Two Things Only

From do-things-that-dont-scale and YC advice:

  • Pre-PMF, focus on exactly two things: writing code and talking to users
  • Everything else is a distraction: conferences, press, partnerships, premature hiring
  • “Nothing is truly finished till it’s released” — ship, don’t polish

How to Choose What to Focus On

The One Metric That Matters

At any stage, identify the single metric that best reflects progress:

StageFocus MetricWhy
Pre-PMFUser conversations / weekLearning speed determines PMF speed
Finding PMFSean Ellis score (% “very disappointed”)Direct measure of product-market-fit
Post-PMFRevenue growth rateProves the model works and scales
Scalingunit-economics (LTV:CAC)Proves growth is sustainable

Post this metric visibly. Altman: Airbnb’s founders posted growth graphs everywhere.

The “Hell Yes or No” Filter

For every opportunity, ask: “Is this a hell yes?” If it’s not, it’s a no. Applied to:

  • Features to build (does this directly improve the focus metric?)
  • Meetings to attend (will this change what we do this week?)
  • Hires to make (is this person a barrel who unblocks a bottleneck?)
  • Partnerships to pursue (almost always no — PG and Altman both warn against early partnerships)

The Eisenhower Matrix for Startups

UrgentNot Urgent
ImportantDo it NOW (customer emergency, server down)Schedule it (product roadmap, hiring, strategy)
Not ImportantDelegate or batch (most email, most meetings)Eliminate (conferences, press, “networking”)

Most founder time goes to urgent-but-not-important. The best founders protect time for important-but-not-urgent.

Focus Anti-Patterns

  1. The “we can do both” trap: You can’t. Choose one, nail it, then consider the other.
  2. Conference tourism: Every hour at a conference is an hour not building. Seibel: skip unless it’s for customer acquisition.
  3. Premature optimization: Optimizing what doesn’t matter yet (infrastructure before PMF, hiring before sales).
  4. Inbox-driven days: Reacting to incoming requests instead of proactively pursuing priorities.
  5. Feature creep: Every “small” addition compounds into complexity that slows everything.
  6. Fundraising as full-time job: Altman: designate one founder to fundraise so the company keeps operating.

Focus Changes by Stage

StageFocus OnIgnore
IdeationProblem validation, customer conversationsProduct, code, design
Pre-PMFCode + users. Nothing else.Press, partnerships, hiring, conferences
Post-PMFGrowth metric + unit-economicsFeatures users aren’t asking for
ScalingThe bottleneck (hiring? infrastructure? distribution?)Everything that isn’t the bottleneck

The Meta-Lesson

Focus is a subtractive discipline, not an additive one. The question is never “what should we add?” — it’s “what should we stop doing?” Rabois’ editing metaphor: the leader’s job is to remove, simplify, and clarify until the team can execute without confusion.

The founders who win are not the ones who do the most things. They’re the ones who do the fewest things, but do them with relentless intensity.

See Also

Sources