UNDERCOVERLABIf you’ve made it this far, you already have an idea in your head. And the first question is always the same: how much will it cost me. The honest answer is that it depends. But "it depends" is no use to you, so let’s get specific. We’ll explain how the prices really come together and where most people get stuck along the way.
Why two quotes for the same thing can range from €5,000 to €200,000
You search "how much does an app cost" and you find everything. It’s not that anyone is lying. It’s that nobody is talking about the same product. The price comes down to three things, and all three can be tiny or enormous.
- How many features the app actually does.
- How many platforms it has to run on (iOS, Android, web).
- The weight of what the user doesn’t see: server, database, accounts, notifications.
That third part is the one that surprises everyone. On a simple app it can be 40% of the cost. When geolocation, chat and payments come in, it climbs to 60%. You’re paying mostly for what you can’t see.
Three real brackets, drawn from our day-to-day
These aren’t fixed rates, they’re orders of magnitude so you can place yourself before sitting down to talk.
| Investment | What’s included | Examples |
|---|---|---|
| Under 15,000€ | Basic iOS and Android MVP: sign-up, listings, simple geolocation, booking without payment, design. | Local marketplace, services app with bookings, directory with a map. |
| Up to 30,000€ | Full MVP: notifications, basic chat, scanner, web panel, structure ready for payments. | Management app with dashboard, marketplace with reviews, sports app with live coverage. |
| Over 30,000€ | Complete product: payments, AI, third-party integrations, multilingual, analytics. | SaaS platform, app with streaming, ERP integration. |
Start small, even if it galls you
MVP means minimum viable product. The smallest version you can already put in front of real users. It works, it does one thing, and it does it well.
Most of the features you currently consider essential aren’t. The client arrives with twenty ideas and six come out that add value on day zero. The rest are assumptions. Do the minimum, launch, listen and decide where to invest afterwards, when you have data instead of hunches.
Set aside between 20 and 25% of the budget for what you hadn’t foreseen. Real users always ask for things that come up in no meeting.
The apps that try to do everything end up doing nothing
Think of the apps you use every day. There’ll be ten or fifteen, and each one does one specific thing very well. The ones that tried to do everything, you’ve already deleted.
If you turn up with three features that have nothing to do with each other, the best advice we can give you is to pick one.
Pick one, make it watertight, validate it. Then we’ll add the rest with judgement.
How we put the quote together
We usually give you two options so you can see the difference with the numbers in front of you.
- Option A: pure MVP. Sign-up and the core feature, nothing else.
- Option B: what you already know you’ll need but isn’t critical on day zero: notifications, scanner, reviews, web panel.
The difference is usually 60 to 80% more cost and four to six more weeks of work. Our recommendation is almost always Option A. And we work in milestones, you don’t pay it all at once: a portion at the start, one on delivery of the design, one with the app in testing and the last one a month after going into production. That way both sides are covered.
What if I build a website first to test?
You can, but you need to know what you’re buying. A landing page to collect emails and measure whether the idea is of interest costs little and works very well. A website with real features (sign-up, listings, bookings) isn’t much cheaper than the app, and on top of that what you build for web isn’t reused for the app because they’re different technologies.
What we do recommend: a simple landing page while we build the app, to start building a user list from day one.
Has AI made it cheaper? Yes, with small print
We use AI to generate code, prototype and automate tests. That lets us offer prices that were impossible two years ago. So far, good news.
The small print: AI writes code, but someone has to review it. And not just to see whether it works, but whether it’s secure and whether your users’ data is protected.
There was a case on LinkedIn that says it all. Someone was boasting about having built their platform with subscriptions and payments without writing a single line of code. The first comment was from a security firm showing them that all the users’ data, names, emails and payment details, was accessible to anyone. In Andorra, with data protection law on top, an app like that can land you a fine that’s a hundred times what doing it properly would have cost. AI is a tool, not the engineer.
Eight questions before asking for a quote
If you arrive at the meeting with these answers clear, we both save two weeks.
- What specific problem the app solves, in one sentence.
- Who has to use it.
- What the core feature is that’s already useful on day zero.
- iOS, Android or both.
- Whether you have the legal texts: privacy and terms.
- What total budget you have, first-year maintenance included.
- How you’ll get your first users.
- Whether you’re willing to launch minimal and keep iterating.
Where we come from
We’re an Andorran company that’s been building apps and platforms for years: sports clubs, marketplaces, verification tools, media and local and international businesses. The first step isn’t a quote, it’s a no-strings conversation to see whether your idea stands up.
