You have an app idea. Now what?

Most app ideas die in the first week, and not because they were bad. They die because the first week gets spent on the wrong things: a name, a logo, a quote from an agency, a search for “a developer”. None of that tells you whether anyone wants the app. Here is a first week that does, and it costs nothing.

1. Write the job before the app

An app is a tool for a job. Write the job as one sentence, in the words of the person who has the problem: “Parents at our swim club need to know which lane their kid is in before they leave the house.” Not “a community app for sports clubs”. The second sentence sounds like a product. The first one tells you exactly who to talk to and what the first screen has to show.

The trap: writing the sentence about features (“lets users book, pay and chat”). Features are your guesses about the solution. The job is the thing that stays true even if every feature changes.

2. Draw five screens, badly

Take a pen. Draw the five screens someone would see on the way to getting the job done, as boxes with a few words in them. The first screen, the one where they do the main thing, the one where they see the result, and whatever two sit in between. Arrows for where each tap goes.

Badly is the point. A neat drawing invites comments on colours. A rough one invites the comment you need: “where do I find last week’s?” The book Sprint, from the team at Google Ventures, built a whole week-long method around this: a realistic fake before any real code, so the flow gets judged before the pixels do.

If drawing on paper feels like a waste because you’ll redo it, start in the App Builder instead. Pick the starter closest to your idea, rename it, and you have five real screens you can tap through on your phone in a few minutes, with the arrows already drawn.

3. Show it to five people who have the problem

Not friends, not family, not “people in general”. Five people the sentence in step 1 describes. Hand them the drawing or the phone and ask them to do the job out loud. Then stop talking.

You are listening for two things. Where they hesitate, which is where a screen is wrong. And whether they ask “when can I have this?”, which is the only compliment that counts. If nobody asks, that is not failure. It is a week’s worth of information that an agency would have charged you for after the build.

4. Know the three costs before you build

The dream gets expensive at exactly three points, and all three are public:

• Apple: $99 a year for a developer account, and a Mac with Xcode to build and upload an iPhone app. There is no way around the Mac for the upload step, whatever you read.
• Google: $25, once. New personal accounts must run a closed test with at least 12 testers for 14 days before the app can go public. Recruit 20, because the clock resets if you drop below 12.
• Your evenings. The honest answer to “how long” from people who have done it is a few months of evenings and weekends for a first version, not a weekend.

Everything else, including designing the app and seeing it run on your own phone, costs nothing today.

5. The first version is the one you can show

Your goal for week one is not an app. It is a version someone can hold and react to. With a drawing, that is a sketch on a phone screen. With a design tool that renders real screens, it is the app itself, before a line of code exists.

Pick the starter closest to your idea, change the name to your own, and send the preview to the five people from step 3. Their reaction decides week two.