How to Prioritize Your Time as a Startup Founder

Founders rarely run out of things to do. They run out of the right hours to do them in. A day full of Slack, calls and “quick” tasks can feel productive and still leave the one thing that matters untouched.

This guide on how to prioritize your time as a startup founder draws on three sources I read in full: Paul Graham’s essay “Maker’s Schedule, Manager’s Schedule,” a Y Combinator lecture by partner Adora Cheung on prioritizing time, and First Round Review’s piece in which Levels co-founder Sam Corcos shares five years of time tracking. I will mark which advice is theirs, which is a single founder’s experience and which is my own suggestion.

Why founder time disappears

Paul Graham describes two kinds of schedule. The manager’s schedule is the traditional appointment book, cut into one-hour intervals, where you change what you are doing every hour. The maker’s schedule is for people who make things, like programmers and writers, who prefer units of at least half a day. In his words, an hour is “barely enough time to get started.”

A founder usually has to live on both. Graham’s point is that a single meeting on the maker’s schedule can blow a whole afternoon by breaking it into two pieces, each too small to do anything hard in. He also notes a cascading effect: if you know the afternoon will be broken up, you are slightly less likely to start something ambitious in the morning.

This is an essay from 2009, based on his own experience and not on a study. But if you have ever lost a day to three “short” calls, it will sound familiar.

Step 1: Separate real progress from fake progress

Adora Cheung’s lecture starts with a filter. Real startup progress is work that moves your primary KPI, which she says is almost always revenue or active users, and she says you should set weekly goals for it. In the early stage, she says that work almost always takes the form of two things: talking to users and building the product.

Fake progress is work not directly tied to that KPI. Her examples from founder weekly updates: attending conferences, going after awards, networking events and optimizing the wrong metrics. She does not say these are never worth doing. She says that spending significant time on them is almost always a bad idea.

Try the filter on your own calendar. For every item this week, ask: does it help me talk to users, build the product or hit this week’s number? If not, it is a candidate to cut. That is my phrasing of her test.

Step 2: Find where your week actually went

Cheung suggests an experiment: journal the past week hour by hour, noting what you did, what impact you expected before doing it and whether it moved your KPI. She says you will likely be surprised how much was low-value work, and that the cause is not laziness. People are on autopilot, and the natural instinct is to go for the easy, quick task because checking things off feels good.

Sam Corcos takes this much further. He tracks his time obsessively, and in the First Round piece he says that when every minute is accounted for, there is “no room for self-deception” about where time goes. You do not need minute-by-minute tracking for this to be useful. A week of honest hourly notes is enough to start.

Step 3: Keep an idea list, but do not act on every idea

One of Cheung’s most practical points: write every new idea in a spreadsheet and do not do it right away. She says switching to whatever you just thought of, which always sounds better now than later, causes whiplash and is a primary culprit in making no progress during the week.

Once a week, go through the list and grade each item:

• Impact on this week’s goal: high (likely to help you hit it), medium (maybe) or low (unlikely).
• Complexity: easy (under a day), medium (one or two days) or hard (many days, may not finish this week).

Her rule is to do high-impact, easy tasks first, then high-impact, medium tasks. Skip anything low impact and hard. She also warns against trying to do everything at once: pick enough tasks that you can complete and do well.

Cheung gives a SaaS example where the top item is visiting ten warm prospects in person and the bottom items are programming tasks. She says a common mistake technical founders make is building first and talking to users later, and that honest grading forces you to go talk to users. That example is her illustration, not data.

Step 4: Protect long blocks for the work only you can do

Back to Graham. His main fix is to cluster the manager’s-schedule work into one place. At Y Combinator he uses office hours, several times a week, set at the end of his working day so they are “never an interruption.” He also describes an earlier trick from his own startup in the 90s: programming from dinner until about 3 am, then doing “business stuff” after waking. Two workdays in one.

You do not need to copy the 3 am schedule. The principle is to keep one long block for deep work and put meetings elsewhere. A hypothetical week:

• Mornings, Monday to Thursday: no meetings. Product or user work only.
• Tuesday and Thursday, 3 to 5 PM: all calls, investor updates and “grab coffee” requests.
• Friday afternoon: weekly review of the idea list.

This is an illustration, not a prescription. Your hours will depend on time zones, team size and investors.

Corcos describes a similar habit: batching his meetings so there is more open space in the day, and blocking time for strategy and writing.

Step 5: Decline speculative meetings politely

Graham explains why speculative meetings, such as “let’s grab coffee,” are nearly free on the manager’s schedule and very costly on the maker’s. His two options were to lose half a day or risk offending someone. His suggested third option was to explain the two schedules, and to ask people to understand the cost.

A line you could use: “I keep mornings for building. Could we do Thursday after 3?” That wording is mine, not his.

Step 6: Watch what gives you energy

Corcos makes a claim that is worth reading carefully. He says that in his experience burnout is only loosely correlated with hours worked and more closely tied to working on things that drain your energy. He reports working 50 to 110 hours a week over five years without seriously approaching burnout, and says he is diligent about spending time on energizing work.

This is one founder’s self-reported experience. It is not research, and it does not mean long hours are safe for everyone. The useful takeaway is the reflection habit: notice which tasks drain you, and ask whether they can be batched, delegated or ended. He also says it is fine to grind through draining work when it is a finite task.

Step 7: Stay close to the product

Corcos also warns about a trap. Around the one-year mark he stopped writing software and shifted to growth, operations and recruiting. He calls the roughly two-year break from the codebase “a costly mistake that was quite painful to undo.” Your role will differ by company, but the lesson applies more widely: before you hand off a core area, check whether you are giving up contact with something that drives the business.

A simple weekly rhythm

1. Monday: Set one goal for your primary KPI for the week.
2. Monday: Grade your idea list by impact and complexity. Choose a few tasks.
3. Daily: Protect a long block for the top task. Write new ideas down instead of acting on them.
4. Mid-week: Cluster calls and meetings into set windows.
5. Friday: Look back at the week. What was real progress? What was not?

Doing the long block

A protected block only helps if you use it. If you find it hard to start, working alongside someone else can help. Refocus pairs you with a partner for a 25, 50 or 75-minute focus session with a shared timer. You each say what you will finish, work in parallel and check in at the end.

This is general productivity advice drawn from the sources named above. Your stage, team and market will change what the right priorities are. Start with one number for this week and one protected block. If you want company for it, start a free session on Refocus.

Cover photo: Unsplash.