“Break it into smaller steps” sounds obvious until you are staring at a task you do not want to do. The usual breakdown often stays too abstract: research, plan, organize, draft, revise. Those are phases, not first actions. Each one can still contain dozens of decisions.
A two-minute first step has a narrower job. It should reduce the distance between you and the work. It does not need to create impressive progress. It needs to make the next move clearer than it was before.
1. Use five tests for a good first step
Before you use a step, check it against these five questions:
- Is it physical? Your hands can open, place, type, copy, circle, gather, or send something.
- Is it observable? You will be able to point to evidence that it happened.
- Is it in one context? It does not require switching among several apps, rooms, or people.
- Is it small enough? You can reasonably reach the boundary in about two minutes.
- Does it leave a trace? The result gives you something concrete to continue or return to.
For example, “think about the presentation” is not observable and leaves no reliable trace. “Open a blank slide deck and type a rough title” passes all five tests.
2. Find the first artifact
Most tasks eventually produce or change something: a document, message, form, pile, spreadsheet, booking, sketch, or conversation. Identify that artifact before you choose the action.
Ask: What object would prove this task is no longer untouched?
- For a report, it might be a document with one rough heading.
- For taxes, it might be a folder containing the first statement.
- For a difficult email, it might be a draft with the recipient and subject filled in.
- For cleaning, it might be one cleared surface or one category of objects gathered together.
- For making an appointment, it might be the provider’s contact page open with your calendar beside it.
Choosing the artifact prevents the step from turning into vague preparation. “Get ready to work” can expand forever. “Put the brief and a blank sheet side by side” has a clear result.
3. Write the step with a verb, object, and boundary
Now turn the artifact into an instruction your hands can follow:
[Physical verb] + [specific object] + [small boundary].
Open the budget sheet and label the first empty column. Put every mug on the counter beside the sink. Type one rough sentence under the proposal heading.
Use a physical verb
Prefer verbs that describe a visible movement: open, place, copy, type, highlight, gather, label, photograph, or send. “Decide” can work only when you also state how the decision will become visible, such as “circle one of the three dates.”
Name the exact object
“Open the file” makes you decide which file. “Open the Q3 planning document” removes that decision. If you do not know the exact object, your first step may be to locate it: “Search Drive for ‘Q3 planning’ and open the newest result.”
Add a stopping boundary
Boundaries make the action safe to enter. Use one sentence, one heading, one item, one surface, three files, or the first two minutes. A boundary is better than “until I make good progress,” because you can recognize when it is complete.
Lower the first-pass quality
If the task carries a high standard, state that the first output may be rough: one placeholder heading, one clumsy sentence, three unfiltered ideas. The first step should create material to improve, not demand the final answer immediately.
4. Compare vague tasks with usable first steps
Prepare a client presentation
Too broad: Make an outline.
Two-minute first step: Open a blank deck and type three questions the client needs answered.
Deal with taxes
Too broad: Organize my documents.
Two-minute first step: Create a folder named “Tax documents” and move the first bank statement into it.
Reply to a difficult message
Too broad: Figure out what to say.
Two-minute first step: Open a draft and write one sentence that states the outcome you want from the conversation.
Clean a crowded kitchen
Too broad: Start tidying.
Two-minute first step: Put every cup and mug beside the sink, then stop.
Apply for a role
Too broad: Update my résumé.
Two-minute first step: Copy the job description into a note and highlight the first repeated skill.
Notice that each action creates a starting artifact. None of them requires motivation for the entire task. Each removes at least one decision from the next session.
5. Troubleshoot a step that still does not start
It contains “and”
“Open the report and review the notes and write the introduction” is three steps. Keep the first clause. When it is done, choose again.
It depends on missing information
Make requesting or locating that information the first action: “Message Priya: ‘Which date should the forecast use?’” or “Open the contract and highlight the renewal date.”
It begins with research
Research has no natural edge. Add a question and a boundary: “Search for the current filing deadline and copy the date plus source into the note.” Do not ask yourself to “research taxes” for two minutes.
It is physically small but emotionally heavy
Reduce the commitment, not just the size. You might write a private draft that will not be sent, put the phone number into your contacts without calling, or ask a trusted person to sit nearby while you begin. A step can be mechanically simple and still feel difficult.
It feels pointless because the project is huge
Choose a step that changes what tomorrow-you will encounter. Leave the right file open, write the question you need answered, or place the materials together. A useful trace makes the next entry less costly even when the project remains large.
6. Decide what happens when two minutes end
When the timer ends, pause. Do not automatically turn a small commitment into an unlimited one. Look at the artifact and choose:
- Stop and leave a marker: write the next physical action at the top of the page.
- Run another bounded step: choose one more sentence, item, or short timer.
- Schedule a return: pair a specific time or cue with the action you will take next.
If you continue, it is because you made a new choice, not because the original two-minute promise was dishonest. If you stop, the first step still succeeded: the task now has a visible edge you can pick up later.