Source: https://www.hype-stack.dev/stacks/how-do-i-make-an-app

# How Do I Make an App? Four Routes, Ranked by Cost

> The first decision is not which tool. It is whether the thing you are imagining needs to be an app at all, because the
> answer is often no and that saves you months.

If you have an idea and no clear route to building it, the useful order is: work out what kind of app it is, pick the
cheapest route that fits, and get something in front of a real person before spending money. This page walks that,
answers first.

We sell source code you clone and run yourself, which is one route among several and not the right one for most people
asking this question. The product section is at the end.

## Start by working out which kind of app you mean

**Does it need to be on a phone's home screen, from an app store?**

- **No, people can open a link:** Build a web app. It is far cheaper and faster.
  - One codebase, works on every device with a browser, updates instantly with no store review
  - Can still be installed to a home screen and work offline, without being in a store
  - This covers most business ideas, internal tools, and marketplaces
- **Yes, it must be in the stores:** Then you are building a mobile app, and the cost changes.
  - Two platforms, store review on every release, and developer accounts with annual fees
  - Justified by device features, background behaviour, or an audience that only looks in a store
  - Cross-platform frameworks let one codebase serve both, at some cost in polish

_Most people describe an app and mean a web app they can send a link to. Establishing this first removes most of the cost._

## The four routes, cheapest first

|  | No-code builder | AI app builder | Template plus a developer | Built from scratch |
| --- | --- | --- | --- | --- |
| Who can do it | You, with no code | You, describing what you want | A developer | A developer or team |
| Time to something usable | Days | Hours to days | Months | Weeks |
| Where it stops | The platform's limits become yours | The second half of a real product | Whatever you are willing to build | Nowhere, at a price |
| Do you own it | No, it lives there | Sometimes; check the export | Yes | Yes |

The ownership row is the one that catches people later. Something built inside a platform is difficult to move, and the
time to discover that is before you have customers depending on it, not after.

## Validate before you build anything

The most common expensive mistake is building for six months and then finding out. Two cheap steps first.

Describe the thing to five people who would supposedly use it, and ask what they do today instead. If the answer is a
spreadsheet they are content with, that is worth knowing now.

Then fake it. A form, a shared document, and you doing the work manually will test whether anyone wants the outcome.
Plenty of successful products ran that way for their first months, and the version you eventually build is better for it
because you learned what actually matters.

## What making an app actually involves beyond the idea

Whichever route you pick, a real application needs the same supporting parts: a way for people to sign in, somewhere to
store data, rules about who can see what, a way to take money if it is a business, somewhere to look when something
breaks, and someone to answer emails when a user is stuck.

The interesting feature is usually a small fraction of that. This is not a reason not to build, it is a reason to choose
a route that gives you the boring parts rather than making you write them.

## About making money from apps

A common follow-on question. Being direct: most apps earn very little, and the ones that earn do so because they solve a
specific problem for people who already have that problem and money to spend on it.

Building first and looking for an audience afterwards is the pattern that fails most reliably. If revenue is the goal,
start from a group of people you can already reach.

## Where this stack fits

If you have concluded that you want a web application, that you want to own the code, and that you or a developer will
build it, this is one starting point among several. What it provides is the supporting layer above: sign-in, accounts,
organizations, roles and invitations, a Hono backend with Prisma and Kysely on Postgres, a React frontend, an admin app,
and Stripe billing, plus editor rules that give a coding agent the conventions of the codebase.

What it does not provide is your idea. The templates are shapes to start from rather than finished products, and the
part that makes it yours is the part you write.

If you are at the validation stage, do not buy anything yet. Fake it first. The templates and packs below are for when
you know what you are building.



Each route has a longer page: [How to Build a Web App, Decision by Decision](/stacks/how-to-build-a-web-app) and
[Best AI App Builder 2026](/stacks/best-ai-app-builder), and the website-first case in
[How to Create an App for a Website](/stacks/how-to-create-app-for-website). For the phone specifically, start with
[How to Create a Mobile App From a Website](/stacks/how-to-create-mobile-app-from-website).
