What to build first, if you cannot code

Build something you personally need this week, that one person will use, and that would still be useful if it looked plain. Projects fail for lack of a clear finish line far more often than for lack of technical ability, and a first project with an audience of one has a very clear finish line.

What tends to finish

• A tool that replaces a spreadsheet you already maintain by hand.
• A booking or enquiry page for something you already do.
• An internal list that two or three people need to see the same version of.
• A one-page site for a thing that currently exists only in a message thread.

What tends to stall

Anything with accounts and permissions on day one. Anything where the value depends on other people showing up. Anything you cannot describe in two sentences. None of these are too hard to build; they are too hard to know when you have finished.

The order that works

1. Describe the thing in two sentences. If you cannot, that is the actual first task.
2. Build the plainest version that does the job, with no accounts and no settings.
3. Use it yourself for a week. Fix only what genuinely annoyed you.
4. Put it online and give it to one other person.
5. Only then decide whether it deserves to be bigger.