Every solo founder I know has a to-do list with forty defensible items on it. Fix the onboarding bug, write the case study, reply to the churned customer, refactor the billing code, DM the warm intro from last week. Every single one is a reasonable use of an hour. That is exactly the problem: when everything is reasonable, nothing tells you what to do first, and you end up doing whatever feels most urgent in the moment instead of whatever actually matters.
A time budget doesn't solve a ranking problem
Blocking hours for marketing or product work fixes how much time you spend on a category. It does not fix which task inside that block you actually do. I can protect six hours a week for distribution and still burn all six of them on the wrong six hours: polishing a blog post nobody asked for while a warm reply from a prospect sits unanswered for three days. The scarce resource for a solo founder was never time. It was a rule for choosing between two things that both look worth doing.
The three-question test
Before I open my laptop, I run every candidate task through three questions, in order, and stop at the first one that gives a clear answer.
1. Is a specific person waiting on this right now? Not "someone will eventually need this." A specific, named person: a customer with an open ticket, a prospect who replied yesterday, a user stuck mid-onboarding. If yes, that task wins, full stop, regardless of what else is on the list. Nothing compounds distribution or product quality faster than someone concluding you don't respond.
2. If I skip this for two weeks, does the cost show up or stay invisible? Some tasks have a cost that shows up immediately if delayed: a decaying deal, a support queue, a broken signup flow. Others have a cost that stays invisible for months even if you never do them: a nice-to-have integration, a rebrand, a competitor feature you don't actually need. Anything in the second bucket gets moved to a "someday" list, not today's list, no matter how appealing it is to work on.
3. Am I the only person who can do this, or am I doing it because delegating feels like more work than doing it myself? This is the question that catches the trap most solo founders fall into without noticing: doing a task yourself not because you're the only one who can, but because writing instructions for a contractor or tool feels slower than just doing it. If the honest answer is "someone else could do this at 70% of my quality for a fraction of my time," it does not belong on today's list either, even if nobody else is currently doing it. That's a hiring or automation problem to solve this month, not a task to keep re-doing yourself every week.
Whatever survives all three questions is what gets done today. Everything else goes on a list you review weekly, not daily, so it stops competing for your attention every morning.
A worked example
Say your Monday list has: reply to a customer who hit a bug (question 1: yes, named person, waiting) — write this week's blog post (question 1: no; question 2: invisible cost if delayed two weeks) — fix a flaky test suite that only you understand (question 1: no; question 2: cost shows up eventually as slower shipping, but not this week; question 3: you're currently the only one who can touch it, which is itself the real problem to flag) — update the pitch deck for an investor call next month (question 2: invisible cost, plenty of runway before it matters).
Only the customer bug clears question 1 outright. The blog post and the deck both get pushed to the weekly list. The flaky test suite is the interesting one: it doesn't win today's list, but it flags a real gap, since being the only person who can touch a piece of the codebase is exactly the kind of bottleneck this test is designed to surface before it becomes a crisis.
Where the test breaks down
The test assumes you're honest about question 1. Founders under-report how often "someone is waiting" applies, because product and content work feels more like "real founder work" than a support reply does. It also assumes you actually keep a weekly list to send the losers to. Without one, everything that doesn't win today's test just evaporates instead of getting revisited, and you end up rebuilding your priority list from memory every single morning, which is its own tax on the day.
The other failure mode is running the test on too many items. If your list has forty things on it before you start, the test itself becomes the time sink. Cap the candidate list at five or six items before you run the questions. If you can't get it under six, that's a signal you're avoiding a bigger decision about what to stop doing altogether, not a signal that you need a better filter.
Quick answers
Does this replace a weekly time budget for marketing or product work? No. A time budget tells you how many hours go to a category this week. This test tells you which specific task inside that block to do first, on any given day.
What if two tasks both clear question 1? Take the one with the shorter time-to-resolution. Clearing one fully-closed loop beats making partial progress on two.
How often should the "someday" list get reviewed? Weekly, same day and time, ideally the same sitting where you plan the coming week. Anything still sitting there after a month either gets scheduled for real or deleted, since a list you never trim stops being useful as a filter.
Pick five items off today's list, run them through the three questions once, and notice how few actually survive to the top. That gap between what felt urgent and what the test kept is usually the actual size of your time management problem.