How to Start When You Don't Know Where to Start
The missing-first-step test
There is a particular kind of stuck that feels confusing, because the task itself is not confusing. You know what needs doing and could explain it in a sentence. Yet when you sit down to begin, nothing moves — not because you are unwilling, but because no first move presents itself. The task is a destination with no doorway.
Try this test before you do anything else. Finish the sentence: “I will start by …” — with a physical verb and a result you could see when it is done, in ten minutes or less. “I will start by opening the document and typing the first heading.” That passes. “I will start by working on the report” does not: there is no verb you can actually perform, and no point at which you could say it is done. If you cannot finish the sentence that way, you do not yet have a first step. That is the whole problem — and it is a much smaller problem than the task itself.
This is different from a task that is simply too big, and different from a task you are avoiding. A big task still has an obvious first move; you just dread the size of the rest. An avoided task has a first move you can name but do not want to take. Here, the first move is genuinely missing. The fix is different: you need one visible move.
Why breaking it down is not quite the answer
The usual advice is to break it down. It often fails here because most breakdowns stop one level too high. “Write the report” becomes “research, outline, draft, edit” — and every one of those is still a destination, not a move. You can stare at “research” for just as long as you stared at the whole report, because “research” does not tell your hands what to do.
A useful breakdown ends in a move, not a phase. A phase describes a kind of work; a move describes an action with a visible result. “Research” is a phase. “Open three articles and copy one useful sentence from each into a note” is a move. The test is whether, after ten minutes, something exists that did not exist before: a heading, a sentence, a list, a sent message. If nothing visible has changed, you have not started.
So do not ask, “What are the parts of this task?” Ask, “What is the smallest thing I could do that would leave a visible trace?” That question is stricter, and that is why it works. It refuses every answer that is really just the task wearing a smaller name.
Cut until the first move is visible
Start with the task as you would say it, then cut it three times, each cut turning something abstract into something you can see.
First, swap every noun for a verb. “The presentation” becomes “make slides.” “The kitchen” becomes “clear the counter.” If a step has no verb in it, it is not a step yet. Second, shrink the time until it fits in ten minutes. Not “write the introduction” but “write the worst first paragraph.” Ten minutes is short enough that the move feels cheap to attempt, and long enough that something real can appear. Third, name the visible result in advance: “one heading on the page,” “five items in a box,” “one email sent.” If you cannot name what done looks like, cut again.
Keep cutting until the move feels almost embarrassingly small — small enough to do without deciding about the whole task. You are committing to one visible thing, for ten minutes, with permission to stop. Most people do not stop, because starting was the hard part.
- Verb, not noun: “make the first slide,” not “the presentation.”
- Ten minutes, not an afternoon: a move you can finish before the doubt catches up.
- A visible result: a heading, a sentence, a list, a sent message — something that did not exist before.
Three tasks, three first moves
The first move is almost always smaller than the task suggests. Take “write the report.” The missing step is not the argument or the structure; it is the first mark on the page. The first move is: open the document and type the ugliest possible first heading. It does not need to be right. It needs to exist, because a page with a heading is a page you can edit, and a blank page is a page you can only dread.
Take “sort out the spare room.” There is no first move in “sort out” — it is a whole afternoon disguised as an action. The first move is: pick one surface, the top of the dresser, and clear it completely. The room is not sorted, but it now has a cleared corner proving that beginning is possible.
Take “reply to that difficult email.” The first move is not the reply at all: it is three bullet points, in a private note, of what the reply must say. Once they exist, the email is mostly written — the hard part was deciding what to say, and the bullets already did that.
In each case, the first move does not plan the rest or require you to feel ready. It only creates a visible trace — and the second move usually suggests itself once the first exists.
Where start/ish fits
start/ish is one option built around this exact moment: you unload what is circling in your head — typed or spoken, messy is fine — and it hands back one small, editable next step you can begin in two minutes, with a gentle timer running. When the first step is the problem, that step can be the ten-minute move itself — “open the document and type the first heading” rather than “write the report” — small and visible enough to begin. Afterwards it asks how it went: kept going, made a dent, got set up, or need something smaller? Every honest answer counts.
Free to try, on the App Store for iPhone now, with a 7-day free trial on the yearly plan.