Start with something small and useful

Three common traps

1. 1A pretty page with no job. It looks finished, but there is nothing you would come back to do on it.
2. 2A tool for other people who never asked for it, such as a clever calculator for colleagues. It is hard to know whether it helps, and it usually goes unused.
3. 3A big project at the start. Stories about teenagers building companies with AI make small things seem not worth doing. A large first project is much harder to finish, and an unfinished project teaches less than a finished small one.

What a good first tool looks like

1. 1It handles something that happens every week: expenses, medicines, a reading list, a shopping list, a class timetable.
2. 2It has one job and fits on one screen.
3. 3Its data stays on your computer, so there are no accounts, passwords or servers.
4. 4You can describe it in two sentences.
5. 5You are its first user, so you can tell at once whether it works.

The filter question

1. 1Ask yourself: would I open this tomorrow? If yes, build it.
2. 2If not, finish this sentence: "Every week I have to ____ by hand." The blank is usually a better first project.
3. 3Once the small version works, you can grow it step by step. Bigger tools are built the same way.