Unstick the side project tonight
A first step you can do tonight in under an hour, and a collaborator that keeps you moving every session after.
The problem
The project you care about has been "about to start" for months because the first step is foggy and the whole feels huge.
What you get
A working system with the steps, tools, checkpoints, and expected return made explicit.
- Setup
- 15 minutes
- Back each week
- 3 hours
- Difficulty
- starter
Expected return
The working case
A planning estimate, not a guaranteed result. Measure the first four weeks against your own baseline.
144
hours returned per year
At 3 hours/week across 48 working weeks.
1
week to earn back setup
Compare the setup estimate with the weekly time returned.
Operating contract
Input
The source material, constraints, and examples a human would need to do this work well.
Checkpoint
A person reviews judgment calls, sensitive content, unfamiliar tools, and irreversible actions.
Success signal
Track time returned, corrections required, and exceptions. Keep it only if the measured result compounds.
The steps
- 01
Name the project and the fog.
Copy this prompt
I keep not starting this project: [describe it and why it matters to you]. What is blocking me feels like: [too big / do not know where to start / afraid it will be bad]. Break it into: 1) the smallest real first step doable tonight in under an hour, 2) the first week of tiny steps, 3) what to explicitly ignore until later.
- 02
Do the tonight-step before you sleep. Momentum, not planning, is the deliverable.
- 03
Start each work session by pasting where you left off and asking for the next smallest step; end by asking it to log progress in one paragraph you save.
What it runs on
- Claude
Doubles as planner and working collaborator (naming, drafts, feedback).
Where this goes wrong
- Planning with AI can become the new procrastination; the rule is plan for 10 minutes, then do the step.
- Keep the progress log yourself; chat history is a terrible archive.
Definition of done
Run it for four weeks. Then make it earn its place.
- □ Baseline the manual time before launch.
- □ Keep a human approval step for consequential output.
- □ Record corrections and exceptions, not just successes.
- □ Expand, revise, or retire it after the first review.
Build the system around it
Related workflows
If this one stops working, tell us. Three reports in a month and it leaves the library until a person has looked at it again.