How much does it cost to build an app, web app or MVP?

There is no single price for every product. This guide shows practical ranges, the decisions that move them, and an estimator that makes its assumptions visible.

  • Transparent ranges
  • No personal data
  • Calculated in your browser
How to use it
Choose the product shapeA feature, web app, mobile app, or web and mobile.
Add the first-release scopeAccounts, payments, integrations, maps, AI, or data.
Read the assumptionsThe result explains which choices moved cost and timing.

A useful range starts with the result being delivered.

Screen count says very little. The main user journey, data, access rules, integrations, release platforms, and clarity of the starting point matter much more.

These ranges cover projects I can take on directly when the scope is clear. They apply to clients in Italy or worldwide; the final quote depends on what already exists and the first release being delivered. VAT, where applicable, and third-party costs are excluded.

Four project shapes with different assumptions.

They are not fixed packages. They help establish the likely order of magnitude before a quote.

01

A feature or integration in existing software

EUR 3,000–8,0001–6 weeks

One defined outcome: payments, a private area, an API, an internal panel, or a flow to finish.

Assumption: The project runs, access is available, and the work stays within one clear area.
02

A lean web MVP

EUR 10,000–25,0006–14 weeks

A usable web application with one main journey, the required backend, and deployment.

Assumption: One primary user type, few integrations, and secondary features left for later.
03

A mobile MVP

EUR 15,000–35,0008–18 weeks

A first iOS and Android app with a backend, device testing, and a distributable build.

Assumption: The mobile flow is defined and does not also require a full web app and extensive back office.
04

A web and mobile product

EUR 25,000–55,000+14–30+ weeks

The same journey works across web, iOS, and Android with shared backend and product logic.

Assumption: It should be phased: prove the core value first, then add the remaining surfaces.

Build a first range.

Choose the starting point and only the capabilities needed in the first release.

What are you starting with?

The base includes one journey; add anything else below.

How defined is the first release?
What does it need to include? Optional additions

Unknowns cost more than screens.

These are the areas I check before turning an orientation range into a quote.

01

The main user journey

Who uses the product, what they must complete, and which state needs to persist.

02

The starting point

A working repository is different from code without tests, deployment, or available access.

03

Roles and edge cases

Permissions, account recovery, failures, and intermediate states are part of the real product.

04

External services

Payments, maps, email, and APIs add limits, test environments, and failure behaviour.

05

Web, mobile, and stores

Adding iOS and Android means device testing, builds, store materials, and review.

06

After release

Logs, analytics, cloud costs, errors, and updates matter when people actually use the product.

An estimate is useful only when its boundaries are clear.

The final quote states what I will deliver, what I need from you, and what remains outside the scope.

Normally considered

  • defining the agreed outcome and cases
  • frontend, backend, and selected integration work
  • testing the journeys included in scope
  • deployment or release preparation when agreed
  • the agreed code, access, and handover notes

Estimated separately

  • VAT, subscriptions, and usage-based third-party costs
  • branding, content production, and marketing campaigns
  • migrations or integrations that have not been reviewed
  • ongoing maintenance unless included in the quote
  • new features introduced after the scope is agreed
Mappu brings planning, maps, and on-trip guidance together

Mappu taught me to estimate the product, not just the screens.

Mappu is a product I built and maintain directly. It combines web, iOS, Android, backend services, AI, maps, audio, accounts, and payments—the same elements that move an estimate.

One product, several surfaces

Accounts and data need to remain consistent across web and mobile.

AI inside a real journey

Generation feeds itineraries, maps, and audio rather than living in an isolated demo.

Release and costs continue

Stores, analytics, errors, and usage-based services remain after the first deployment.

Before treating the range as a budget.

Can an app or MVP be built for less?

Yes, if the first outcome is smaller: one journey, one platform, fewer roles, and no unnecessary integrations. A static prototype can cost less, but it is not the same as a usable product.

Why can two quotes be very different?

They often include different work. Design, backend, admin tooling, testing, release, account handover, and support after launch must be explicit before prices can be compared properly.

Does the estimator produce a quote?

No. It produces a range consistent with the choices entered. Before quoting, I review the starting point, responsibilities, edge cases, and exact delivery criteria.

What changes when code already exists?

It can reduce the work or add unknowns. I first check setup, the relevant code, tests, access, and the deployment path. A defined feature can only be quoted after that review.

Does MVP mean an unfinished product?

No. The main journey should work from start to finish. Secondary features are postponed, not reliability, data ownership, or the ability to use the result properly.

Are hosting, AI, and store costs included?

Not unless the quote says so. Cloud, email, maps, AI models, developer accounts, and other services have their own costs. I make them visible because they continue after delivery.

Do you have a first release to define?

Send me the estimator summary and whatever already exists. I will tell you which assumptions need checking before we discuss a quote.