What thirty days is actually for
One finished, documented project. That is the whole goal. Not three projects, not a personal website, not a rebrand of how you present yourself. One thing a stranger can open, understand and question you about.
Thirty days is enough for that if you are working around a job or a job search, which most people reading this are. It is not enough for something ambitious, and pretending otherwise is how people arrive at day thirty with a folder of half-finished pieces and nothing to link to.
Everything you build here lives outside this site. We do not store your work, we do not host it, and there is no account to create. Copy the plan below into whatever you already use, a document, a notebook, the notes app on your phone. It is yours.
In plain English
- Capstone:
- The final piece of work that pulls together everything a course covered.
- Time box:
- A fixed amount of time you allow for a task, after which you stop regardless.
- Artefact:
- The actual thing you produced, as opposed to a description of it.
- Definition of done:
- A short written test that tells you the work is finished.
Roughly ten to fifteen hours across thirty days is the realistic budget for most people. That is about three hours a week. Plan for that, not for the version of you who suddenly has free weekends.
The four weeks
Week one: choose and scope. Mine your own work history for candidates, pick one, and write the definition of done. End the week with a written scope you could hand to someone else, including the list of what you are deliberately leaving out. Two hours is plenty. Do not start building.
Week two: build the ugly version. Get it working end to end, badly. Every step can be manual. The output can be poor. What matters is that it goes from input to something usable, because that is what tells you where the real difficulty is. Keep a rough log of what breaks.
Week three: improve one step, then stop. Pick the single weakest step and make it better. Then check your definition of done and stop building. This is the week where scope creep kills projects, so read your own out list before you add anything.
Week four: document and place it. Write the one page, take the screenshots including one of something going wrong, host it somewhere with a working link, and rehearse the ninety second walkthrough out loud.
Help me turn this into a dated plan I can keep.
My project idea: [describe in two sentences]
Hours I realistically have per week: [number]
Start date: [date]
Produce a four week plan with specific tasks and dates, following this structure:
Week 1 choose and scope, Week 2 build it end to end badly, Week 3 improve one step then stop, Week 4 document and publish.
For each week, give me one measurable outcome and one thing I must NOT do that week. Assume I will overestimate what I can finish.
Checkpoint
Four weeks: scope it, build it badly end to end, improve one step, then document and place it. One project, finished.
The checklist
Copy this and tick as you go. It is deliberately short, because a checklist you actually finish beats a thorough one you abandon.
MY PORTFOLIO CHECKLIST
Scope
[ ] Problem comes from work I actually understand
[ ] I can name who was losing time and roughly how much
[ ] Definition of done written as 3 to 5 testable checks
[ ] Written list of what version one will NOT do
[ ] Any real workplace data replaced with invented sample data
Build
[ ] Works end to end, even roughly
[ ] I can explain every part of it in my own words
[ ] Rough log kept of what broke and what I tried
[ ] I stopped when the definition of done was met
Document
[ ] One page: problem, what I built, what failed, what I would change
[ ] States exactly where AI helped and where I decided
[ ] Before and after described honestly, no invented numbers
[ ] Two or three screenshots, including one thing going wrong
Place it
[ ] Link opens logged out, in a private window, on a phone
[ ] One factual line ready for a CV, 25 words maximum
[ ] Ninety second walkthrough rehearsed out loud
[ ] I have answers for the two hardest questions about it
If you reach day thirty and the project is not finished, do not extend it. Cut a feature, note in the write up why it stayed manual, and publish. A finished project with an honest gap is evidence. An unfinished project is nothing at all, no matter how good it would have been.
Diagnosing where you are stuck
Different weeks fail in different ways. If you are stalled, the useful question is which of these you are actually doing.
โ Weak prompt
Prompt
I am not making progress on my portfolio project. Any tips for staying motivated?
Output
General encouragement about breaking tasks into smaller pieces, setting aside dedicated time, and remembering why you started.
You already know all of this. Motivation is almost never the real problem, and asking about it hides the specific thing that is actually blocking you.
โ Good prompt
Prompt
I am stalled on my portfolio project. Here is my scope, my definition of done, and what I have done so far. Tell me which of these is happening: my scope is too big, I am polishing instead of finishing, I am blocked on one technical step, or I picked a problem I do not care about. Then give me the smallest next action, one sentence.
Output
A direct diagnosis, likely that the scope covers three workflows rather than one, and a next action of cutting two of them and writing the reason in the log.
Naming the failure mode is what unblocks you. The single next action is deliberately small enough that there is no reason not to do it tonight.
After day thirty
If it went well, you now have one thing to show and a much better sense of what you can do. That is the point at which a second project becomes worth considering, and the second one is always easier.
You will also finish this course with the Proof of Work badge. Treat it the way lesson one described. It marks that you did the work here, and it is a reasonable thing to feel good about. It is not the evidence. The link you can now put on an application is the evidence, and it exists because you built something, not because you completed anything.
I have finished my first portfolio project. Here is the write up.
Help me decide what to do next. Answer three things:
- Based on what this project shows, what kind of role is it evidence for?
- What is the single biggest gap between this and the roles I want, listed here: [roles]
- What smaller second project would close that gap in about ten hours?
Do not suggest more courses. Suggest work I could build.
[paste your write up]
Nobody can promise you a job, and anybody who does is selling something. What you can control is whether a stranger looking at your application has something real to look at. Thirty days from now, that can be true.
๐ Quiz
Question 1 of 4What is the goal of the thirty day plan?