How Much Does It Cost to Build a Mobile App in 2026?

In the United States, building a mobile app in 2026 costs somewhere between USD 25,000 for a stripped-down MVP and well over USD 300,000 for a feature-heavy enterprise or marketplace product. Most funded business apps land between USD 80,000 and USD 180,000. The number people hear quoted, whether it is 30,000 or 150,000, almost always comes down to hours multiplied by an hourly rate.

That is worth saying plainly because the quotes do not disagree about anything except scope. One team is quoting a login screen and a static content feed. The next is quoting a chat system, a payments ledger, an admin panel, and two years of support. If you know how many hours are in the estimate and what the rate is, you can compare two quotes that look wildly different on paper.

The rest of this guide breaks the price down by project type, platform, feature set, and team model, then shows you how to build your own budget from the bottom up. Figures are typical US ranges. They shift with regional labour costs, and they move over time as tooling improves.

Table of Contents

• How Much Does It Cost to Build a Mobile App by Project Type?
• What Does Each Platform and Feature Cost?
• Development Team Models and Hourly Rates
• What Affects the Price?
• Where the Hours Actually Go, Phase by Phase
• How to Calculate a Realistic App Budget
• Ways to Save
• Questions to Ask Before You Accept a Quote
• Frequently Asked Questions
• How much does it cost to build a simple mobile app?
• Is it cheaper to build one app for iOS and Android?
• How much should I budget for app maintenance each year?
• Should I hire an app development agency or a freelancer?
• How long does it take to build a mobile app?
• What is the cheapest way to launch an MVP?
• Start With an MVP Budget

How Much Does It Cost to Build a Mobile App by Project Type?

The cleanest way to size a project is to name the kind of product you are actually building. A simple MVP with a handful of screens and no back end is a different budget conversation from a marketplace with payments, chat, dispute handling, and an operations console.

Project type	Low	Typical range	High	Typical timeline	What the range assumes
Simple MVP or proof of concept	USD 25,000	USD 40,000 to 80,000	USD 120,000	2 to 4 months	One platform or a cross-platform build, 8 to 12 screens, basic login, hosted back end, no custom integrations
Standard business app	USD 70,000	USD 90,000 to 180,000	USD 275,000	4 to 8 months	iOS and Android, custom UI design, admin panel, third-party API integrations, analytics, QA across device types
Enterprise internal tool	USD 120,000	USD 150,000 to 300,000	USD 500,000	6 to 12 months	SSO, role-based access, audit logging, security review, legacy system integration, dedicated environments
Marketplace or two-sided platform	USD 150,000	USD 200,000 to 400,000	USD 700,000	7 to 14 months	Payments and payouts, chat or messaging, ratings, dispute workflows, moderation tooling, search
Public-service or civic app	USD 60,000	USD 90,000 to 200,000	USD 350,000	6 to 12 months	Accessibility compliance, public records requests, procurement overhead, content migration, support and training costs

Those are typical US ranges for commercial work in 2026. Regional rates differ, the same scope priced in South or Southeast Asia can land a third to a half lower, and every figure drifts as tooling and rates change.

One pattern shows up repeatedly in developer forums: quoted prices cluster tightly in the USD 50,000 to 150,000 band for a mid-complexity app, and fall apart at both ends once scope is unclear. Very low quotes usually mean a template with your logo on it. Very high ones usually mean native builds on both platforms with a full custom design system and a long support tail.

What Does Each Platform and Feature Cost?

Platform choice sets your baseline. Building two native apps means paying twice for a lot of the same interface and business logic, while a cross-platform build writes most of it once.

Approach	Relative cost	Where it fits
Single native app (iOS or Android)	1x baseline	Test-the-demand projects, internal tools for one controlled device fleet, apps that lean on hardware-specific features
Two native apps	roughly 1.7x to 2x	Products where platform feel, performance, or store ranking justify the second build
Cross-platform (Flutter or React Native)	roughly 1.2x to 1.5x one native build	Most commercial apps with shared UI, standard forms, lists, and API-driven data
Progressive web app	roughly 0.4x to 0.7x	Internal tools, event apps, forms, content catalogues, anything not needing store distribution or offline use
No-code or low-code builder	USD 3,000 to 40,000 a year plus internal time	Simple internal dashboards, prototype validation, straight-line CRUD workflows

Feature work is where a fixed estimate quietly falls apart. These are typical hours, before the rate is applied, for a mid-complexity build.

Feature	Typical hours	Notes
Email and password authentication	40 to 80	Usually the cheapest real feature in any build
Social login	30 to 60	Priced per provider; each one adds setup and testing time
Push notifications	40 to 80	Cheap on iOS, more work on Android where channels matter
Maps and geolocation	80 to 250	Rises fast with geofencing, background location, or route drawing
Payments and in-app purchases	150 to 400	Store rules, receipt validation, refunds, and accounting all add work
Analytics and event tracking	40 to 90	Cheap if planned early, awkward if retrofitted before launch
Chat and messaging	300 to 800	One of the biggest single line items in a consumer app
Offline mode with sync	120 to 300	Conflict resolution is the expensive part
Admin or operations panel	300 to 700	Buyers routinely forget this and hit it late
Localization per language	40 to 100	Plus review time and layout work for right-to-left languages
AI or machine learning features	300 to 1,500	Infers much more than a single screen; budget for evaluation and data work

Back-end work is a separate bill from the app screens. Hosting, databases, API development, authentication services, and third-party SDKs are usually quoted as their own line, and developers consistently flag this as the most underestimated cost in a first quote.

Ongoing service subscriptions also run year after year. Messaging platforms, payment processing at around 2.9 percent plus 30 cents a transaction, push notification services, analytics, mapping calls, and hosting add up to a few hundred dollars a month for a modest app and considerably more for a busy one.

Development Team Models and Hourly Rates

Who does the work matters almost as much as what gets built. The same scope can differ by a factor of three depending on where the hours come from and how accountable the supplier is for them.

Team model	Typical hourly rate	Strength	Weakness
Solo freelancer	USD 40 to 120	Cheapest route to a working first version	Availability, single point of failure, no team coverage for testing or support
Contracted small team (2 to 4 people)	USD 70 to 140	Real coverage across design, front end, and back end at mid-market cost	Less process, fewer specialisations than a full agency
Mid-size agency	USD 120 to 225	Defined process, contract terms, account management, QA discipline	Highest price, and you buy some process you may not need
Enterprise consultancy	USD 200 to 350	Compliance, scale, procurement and audit experience	Slowest, most expensive, overkill for an MVP
In-house team (fully loaded US)	USD 130 to 200	Full control, institutional knowledge, fast iteration later	Roughly six months before real output, plus recruiting risk
Staff augmentation (offshore or nearshore)	USD 60 to 110	Senior skills at a lower cost, scales up and down	You still own project management and quality control

Regional hourly rates explain most of the gap between quotes for identical feature lists.

Region	Typical blended hourly rate
United States and Canada	USD 100 to 200
United Kingdom and Western Europe	USD 90 to 180
Northern and Eastern Europe	USD 50 to 130
Latin America	USD 40 to 80
South and Southeast Asia	USD 25 to 60
Central Asia and Africa	USD 20 to 45

Treat off-the-clock rates as a starting point only. Blended agency rates include sales overhead, account management, and non-billable time that a freelancer absorbs as unpaid work. That gap is why a freelancer at USD 100 an hour and an agency at USD 180 can quote the same number for the same project.

What Affects the Price?

Price follows scope multiplied by rate. When you know which of these multipliers applies to your project, you can adjust someone else’s estimate before you sign anything.

Decision or requirement	Typical effect on the build cost
Second platform, built natively	adds 60% to 100%
Cross-platform instead of two native builds	typically saves 20% to 35%
Tablet and foldable layouts	adds 25% to 50%
Support for older operating system versions	adds 40% to 100%
Rotated and split-screen layouts	adds 20% to 30%
Full custom design system with component library	adds 20% to 40% over template-based UI
Accessibility work to WCAG 2.2 AA	adds 10% to 20%, and is rarely optional for public-sector work
Sector compliance review (health, finance, payments)	adds 15% to 30%
Migrating content from an existing system	adds 10,000 to 60,000 depending on data quality
Compressed timeline, under four months	adds 25% to 50% for parallel work and overtime
Managed content or editorial back end	adds 300 to 700 hours

Two drivers cause most budget overruns. The first is scope creep: change requests made after development starts, each one small, adding up to a rewritten feature set. The second is device coverage, which buyers rarely think about until QA finds a layout problem on a screen shape nobody mentioned.

Public-sector and civic projects carry a different set of costs on top. Accessibility conformance and public records obligations add requirements that a commercial MVP can skip, and procurement timelines stretch a nine-month build across a fiscal year or more. Cities and nonprofits often offset that with grant funding, which is covered further down.

Security requirements are the other place budgets move. A read-only internal app needs far less than one handling health records, payments, or identifiers, and quotes that treat them identically are worth questioning.

Where the Hours Actually Go, Phase by Phase

Buyers repeatedly ask the same question on developer forums: if a build costs 80,000, what is actually in it? This is the standard split agencies use, and it is a useful sanity check on any quote you receive.

Phase	Share of total effort	What it produces
Discovery and product definition	5% to 10%	User flows, feature list, technical approach, acceptance criteria
UI and UX design	10% to 15%	Wireframes, visual screens, a component library, an accessibility spec
Back end, API, and database	20% to 30%	Server logic, data model, integrations, authentication, admin tooling
Front end and app development	25% to 40%	The screens and flows users actually touch, on each platform
QA and device testing	15% to 25%	Test passes, defect fixes, device and OS coverage, store submission checks
Project management and account handling	8% to 15%	Scheduling, status reporting, budget tracking, stakeholder updates

If a quote puts almost everything into development and near zero into QA or project management, it is either a very small project or a quote that will grow once testing starts. Good vendors push back on their own estimates in exactly that situation.

Every stage after launch also carries a running bill. Hosting and databases, messaging, payment processing, analytics, and map services are billed monthly and grow with usage. The Apple Developer Program is a yearly fee, and Google Play charges a one-time registration plus a recurring plan fee. In-app purchase commission is the largest single deduction from digital revenue, and both major stores set the terms.

Marketing and launch budget deserve a line of their own. Agencies and research firms routinely put a first-year go-to-market budget in the same range as a simple MVP build, which surprises founders who had planned for build cost only. Set aside a comparable amount for the year after launch, and treat store optimisation, content, and support staffing as part of it.

Legal, privacy, and compliance work sits in the same bucket for regulated or public-facing products. Terms of service, privacy policy, accessibility conformance, and sector-specific review are rarely included in a development quote unless you ask.

How to Calculate a Realistic App Budget

A budget you build yourself will beat a number someone handed you, because you control the assumptions. This is the method I would use, and it takes an afternoon.

1. Write the user flows, not the features. List each screen a real user passes through, from first open to the moment the task is done. Ten honest flows tell you more than forty feature adjectives.
2. Split must-have from later. Everything needed to launch goes in the first column. Everything else goes in a dated second column so the conversation about scope stops being abstract.
3. Estimate hours per work item. Use the feature table above as a floor, then add 40% for integration, error states, and the testing nobody quotes for.
4. Attach a rate, and check whose it is. Regional blended rates are published widely. Ask each bidder what rate their estimate assumes and what is excluded from it.
5. Add the line items quotes hide. Project management, QA, design, back end, hosting setup, store submission, and post-launch bug fixing each deserve their own number.
6. Add 15% to 20% contingency. Scope grows. That is not pessimism, it is arithmetic.
7. Budget the first year of running costs. Hosting, service subscriptions, store fees, and maintenance are not zero, and pretending otherwise distorts the build decision.

Here is the arithmetic in one line, using the format that a reputable agency will hand you on request. For a cross-platform MVP at 600 development hours, a back end and API at 400 hours, and an admin panel at 180 hours, a blended rate of USD 150 gives roughly USD 177,000 before design, project management, and QA.

That is why the hours times the rate formula is the only number worth arguing about. Two quotes at USD 60,000 and USD 180,000 can both be honest if the first covers 400 hours of template work and the second covers 1,800 hours of custom build. Ask both to show the hours.

Then look at three-year total cost of ownership, because launch cost is the smallest number in the picture. Add annual maintenance at 15% to 20% of the build cost, infrastructure that grows with users, compliance work that recurs each year your sector is audited, and the marketing budget most teams forget entirely. A project budget that only covers the build is roughly half the real requirement.

Ways to Save

Most savings come from scope and sequencing, not from finding a cheaper developer. These are the moves that cut a real budget while leaving the product usable and secure.

1. Launch on one platform first. The single most effective decision most teams make. Shipping on one platform roughly halves the initial build, and the second platform is cheaper to add later once the product has proven itself.
2. Ship an MVP that solves one problem. A narrow first version that people use beats a broad one nobody finishes. Every feature you defer is a feature you do not redesign twice.
3. Use cross-platform frameworks unless you have a reason not to. Flutter and React Native cover the large majority of commercial apps. Go native when you depend on platform-specific hardware, deep system integration, or store ranking advantages.
4. Reuse proven components and SDKs. Authentication, payments, chat, analytics, and crash reporting all have mature third-party options. Building your own is rarely worth the maintenance burden.
5. Decide on integrations before design starts. Adding a payment provider or a municipal data feed after the UI is signed off is one of the most common sources of mid-build change requests.
6. Use open city data and public APIs as your back end. If your product touches transit, air quality, permits, or events, government open data portals often replace a paid data vendor and the integration work that comes with one.
7. Fix the price with a fixed-scope proposal. Hourly billing with an open scope is an invitation to spend your budget. Cap revisions, price change requests separately, and write code ownership into the contract.
8. Plan maintenance from day one. A version that ships with a support commitment and a bug backlog plan costs less over three years than one launched and forgotten.

For civic, nonprofit, and university teams there are funding routes that never appear on an agency invoice. City innovation funds, national digital inclusion grants, hackathon prizes, and volunteer developer pools have all funded first versions of civic apps. A smaller scoped first release is usually easier to fund than a complete one, which is another reason to keep the MVP honest.

One more lever worth knowing: a progressive web app costs roughly half a native build and needs no store submission. For an internal tool, an event app, or a forms-heavy service, that is often the whole answer.

Questions to Ask Before You Accept a Quote

Most bad decisions on this kind of project come from comparing numbers that do not describe the same thing. These questions force a quote to become something you can actually compare against a competing one.

1. How many hours is this estimate, and what rate produced it? A supplier who cannot show both has given you a gut feeling with a currency symbol on it.
2. What is explicitly out of scope? The exclusions list is usually longer than the inclusions list, and it is where the next change request will come from.
3. How many revision rounds per screen or per phase? Unlimited revisions turn a fixed price into an open one.
4. Who owns the source code, and when do we receive it? Ownership should be in the contract, not a promise in an email. Same for design files and third-party licences.
5. What does the price per platform cover? Some quotes include one platform and price the second as an option. That is not obvious until month four.
6. Which devices and operating system versions are included? If the answer is vague, assume the oldest versions your user base still runs, and price accordingly.
7. What happens to the price if we cut a feature? A good quote shows itemised costs so you can remove work and see the reduction, rather than taking it off as a favour.
8. What are the change request rates, and how is the maintenance retainer defined? A retainer with no service levels attached is just a recurring invoice.

Two red flags show up often enough to name. A very low quote for a marketplace or social product usually means a template with your branding swapped in, which no amount of polishing turns into a real product. And a quote that arrives as a single lump sum with no hours attached cannot be audited by anyone, including you.

Frequently Asked Questions

Start With an MVP Budget

The most useful thing you can do this week is define the smallest version of the app that would still prove the idea, and put a number on it. Write the user flows, count the screens, separate must-have from later, and multiply estimated hours by a published blended rate for your region. That gives you a floor you can take to any vendor.

Then ask three bidders for like-for-like quotes and check that each one shows hours, deliverables, revision limits, and code ownership. A quote without those is a guess, and you will pay for the difference later.

Finally, add 15% to 20% contingency and a first year of maintenance and running costs before anyone commits funds. Answering how much does it cost to build a mobile app is not one number; it is scope times rate plus the cost of keeping the thing alive, and the second part is the part most budgets forget.