How to make an app: from idea to App Store

From idea to app: 8 steps

Making an app is like building a house: first you work out who it is for and why, then you draw a plan, and only then do you pick up a hammer. Most problems appear where this order is broken and people "start development" at once.

1. Idea and demand check. Who needs it and are they ready to use it (and pay for it).
2. First version. The minimum set of features that solves the main task.
3. Screens and prototype. What the user's path looks like.
4. Development. Building the app and its back end.
5. Testing. Checking on real phones.
6. Accounts and materials for the stores. Description, screenshots, privacy policy.
7. Publishing. Apple and Google review, then release to the store.
8. Growth. Analytics, fixes, promotion, new features.

What each stage and the app as a whole cost is set out in "App Development Cost". Here we focus on the order of actions and the pitfalls.

Steps 1–2: check the idea and define the first version

The most expensive mistake is to build an app nobody needs. Checking takes weeks, not months, and costs almost nothing.

Check that the problem exists

Talk to 10–15 people from your future audience. Do not ask "would you like this app?" (everyone politely says yes) but "how do you solve this task now, and how much time and money does it take?". If people feel no pain or are already happy with what they have, rethink the idea.

Look at the alternatives

Find 3–5 similar apps in the App Store and Google Play. Read the reviews: what people complain about and what is missing. This is free research of your market in your country that shows where you can do better.

Define the minimum

The first version is one main user task, solved from start to finish. Not "catalog, chat, loyalty, map and rating" but, for example, "book a stylist in three taps". Everything else comes later. Such a version is called an MVP; what it is and what it costs is covered in "What Is an MVP and How Much Does It Cost".

Decide whether you need an app at all

Sometimes a website, an app made from a website (a PWA) or a Telegram Mini App solves the task, and you do not need to spend on the stores. How to choose is covered in "App vs Website". An app makes sense when you need notifications, the camera, geolocation, offline mode or the customer uses the service often.

Step 3: screens, prototype and design

Before the first line of code, draw the user's path. For a first version 5–8 screens are enough: sign-in, home screen, the key action, profile, payment or confirmation.

• Screen map. Boxes on paper or in a free tool: what follows what and where each button leads.
• Clickable prototype. The same screens, but you can tap them. Show it to 5–7 people from your audience and see where they get lost. Fixing a prototype costs hours; fixing a finished app costs weeks.
• Design. Colors, fonts and buttons under common rules. Design is included in our packages; on its own, design and a prototype start at €1,990, and it is a good first step if you have not yet decided whether to go into development (details).

Think ahead about what the stores require: sign-in with the ability to delete an account, a privacy policy, clear explanations of permissions (camera, location). Apple includes these questions in its review rules (App Store Review Guidelines), and adding them at the last moment costs more than planning them from the start.

How long does it take to develop an app

The timeline depends on the number of screens, the back end and integrations. Cubix's overview gives roughly 1–2 months for a simple MVP, 3–6 for a mid-complexity app and 7–12 for a complex one, and these are planning estimates, not rules (Cubix). Our timelines (cross-platform development, one codebase for iOS and Android):

Apple and Google review is not included in these timelines: it is outside our control, and we say so honestly.

What the timeline is made of

• Introduction and brief: a few days. If you have prepared a description and examples, it gets shorter.
• Design and prototype: 1–3 weeks including approvals.
• Development: the longest part, with interim results shown via a link.
• Testing and fixes: 1–2 weeks.
• Store submission: preparing materials and answering review comments.

What pushes the timeline back

Three things: late replies and approvals from your side, new features added "along the way", and waiting for store review. The first one you control fully: name one person who makes decisions and keep the materials ready. We promise timelines only when conditions are met: the brief and materials before the start.

Steps 4–5: development and testing

App development has two parts: what the user sees (the screens) and what works behind the scenes (the server, the database, emails, payments). Clients usually see only the first and underestimate the second.

One codebase or two

You can write separate apps for iPhone and Android, or one codebase from which both versions are built. The second route is faster and cheaper, and for most business apps (orders, accounts, booking, catalogs) the user will not notice the difference. We take the second route. Separate native development is needed for heavy games, complex graphics and some sensors.

Previews by link

A good practice is to get a working build at every stage rather than one final one. You see the app on your own phone while it is still changing, and corrections are cheap. We give preview links for each stage.

What to check during testing

• The main scenario from start to finish on real iPhones and Androids of different sizes.
• What happens without internet and on a weak connection.
• Emails, notifications and test payments.
• Access rights: everyone sees only their own data.
• Launch speed and behavior when the screen rotates.

If you need a contractor for this kind of setup, see how we described the choice in "How to Choose a Web Development Company": the logic of vetting is the same for a site and an app.

How to publish an app on the App Store and Google Play

Publishing is not an "upload" button but a separate process with rules and human review. Start preparing 2–3 weeks before the launch date.

Developer accounts

• Apple: Apple Developer Program, $99 a year (Apple).
• Google Play: Play Console, a one-time $25 payment (Google).

Register both accounts in your own name or your company's, and invite the developer with the rights they need. Then the app stays yours even if you change teams. It is the same principle as with a website's domain and code.

What Google Play needs

If the developer account is personal and recently created, Google requires closed testing before publishing: at least 12 testers for 14 days in a row, and until this step is done, publishing to the store is unavailable (Google Play Console help). Build these two weeks into the schedule. An organization account for a company works differently, so check the rules in the help center.

What the App Store needs

The version you submit for review must be final, fully working and with working links; the app must offer more than a "packaged website" (App Store Review Guidelines, sections 2.1 and 4.2). If the app has sign-in, prepare a test login for the reviewers.

Materials for the store page

• name, short and full description, keywords;
• screenshots for different screen sizes;
• an icon;
• a link to the privacy policy and a support contact;
• answers to questions about the data the app collects.

We prepare these materials, submit the app for review and fix the comments (the "Publishing to App Store and Google Play" service, €390 if you publish outside a package; it is included in the packages). A first rejection happens, usually over small things. We do not promise review times: the stores set them.

Steps 6–8: what to do after publishing

Publishing is the beginning, not the finish. The first weeks show how the app really lives.

• Connect analytics. Installs, retention, funnels: where users drop off. Without it you will be guessing (setup from €590).
• Read the reviews and reply to them. It is free feedback and a trust factor.
• Update. New versions of iOS and Android come out every year, and the app has to adapt.
• Attract users. An app does not collect installs by itself. Store page optimization and install ads start from €590 a month.
• Add features based on data. The next feature is chosen by what analytics and users say, not by gut feeling.

If you want to see how ready you are for such a project in terms of business, team, budget and idea, use the checklist at the top of this article, or write to us on the contact page.

What to choose in your situation

• If you only have an idea so far, start with a demand check and a prototype (design and prototype from €1,990). It is the cheapest way to learn whether to go further.
• If you need a service "like a website, but on the phone", look at a PWA at €2,490: no store publishing is needed and launch is faster.
• If your customers talk to you through Telegram in your city, you may need a Mini App from €3,900 rather than a separate app.
• If you are ready to test the idea on real people, take an iOS + Android MVP from €10,900: 6–8 weeks and publishing to both stores.
• If the app must take payments and be run by a team, take "Business" at €23,900 with payments, an admin panel and analytics.
• If the app has several roles and complex links, start with an architecture session at €3,500. Its cost is credited toward the project.

The scope and prices of an app are on the iOS + Android app page, and the general price list is in the pricing section. We work in eight clear steps from first meeting to support; how it looks in practice is described in How we work.