"How much does it cost to build an app?"
We get this question on every single discovery call. And every time, the honest answer is: somewhere between $5,000 and $500,000. That range is so wide it's almost useless -- but it's the truth. The price of custom software depends on a matrix of decisions, and most founders underestimate at least three of them.
We've built platforms across sports, media, festivals, venture capital, and athlete coaching. Every one had a different budget, a different timeline, and a different set of tradeoffs. Here's what we've learned about what actually drives cost -- and how to think about it before you talk to any agency.
The price ranges by project type
Let's start with the broad strokes. These are realistic ranges for 2026, assuming a competent agency or dev team (not a freelancer on Fiverr, not a Big Four consultancy).
-
Marketing / portfolio site ($5K - $15K): A modern, fast, well-designed website with a few pages, contact forms, maybe a blog. Think WSV -- a venture portfolio site with polished animations and responsive design. No user accounts, no databases, no complex integrations. This is the simplest category, and it's where most businesses should start.
-
Web platform ($25K - $75K): Once you add user accounts, dashboards, content management, or any kind of dynamic data, you've crossed into platform territory. Footy Access is a good example -- a media platform with content publishing, video embeds, and a custom admin dashboard. You're building both a frontend and a backend, and the backend is where complexity (and cost) compounds.
-
Mobile app ($40K - $100K): Native or cross-platform mobile adds another layer. Edge Athlete OS -- a coaching app we built with React Native and Expo -- required offline support, push notifications, real-time data sync, and platform-specific UX patterns. Mobile isn't just "make the website smaller." It's a fundamentally different engineering challenge.
-
Enterprise / multi-product platform ($100K+): When you're building interconnected systems -- think Odunde, which combines a festival platform with a sports component, user management, event ticketing, and admin tooling -- you're in enterprise territory. Multiple user roles, complex permissions, third-party integrations, and the need for the system to grow without being rewritten.
These ranges assume a team that knows what they're doing. If you're getting quotes significantly below these numbers, ask hard questions about what's being cut.
The 7 things that actually drive cost
The difference between the low and high end of each range comes down to specific technical and design decisions. Here are the ones that matter most.
1. Scope breadth
This is the obvious one, but it bears repeating. Every feature you add isn't just development time -- it's design time, testing time, edge case handling, and ongoing maintenance surface area. When we scoped NXTGEN Media as a two-week MVP, the first thing we did was cut 60% of the feature list. That's not because we were lazy. It's because scope is the single largest cost driver, and most founders dramatically underestimate how many features they're asking for.
2. Authentication and user roles
A simple email/password login is straightforward. Add OAuth providers, role-based access control, team management, or multi-tenancy and you've added weeks of development. BookIt Sports needed public users, pool participants, pool administrators, and platform admins -- four distinct permission levels, each with different views and capabilities. That auth layer alone was a significant portion of the build.
3. Real-time features
Anything that needs to update without a page refresh -- live scores, real-time notifications, collaborative editing, chat -- requires WebSocket connections or server-sent events, state synchronization, and conflict resolution. We've built real-time score tracking for BookIt Sports, and the infrastructure required to make that reliable is an order of magnitude more complex than a standard request/response API.
4. Third-party integrations
Every API you connect to is a liability. Payment processing (Stripe), email services (SendGrid, Resend), analytics, social auth, sports data feeds -- each one has its own documentation quality, rate limits, error handling patterns, and breaking change frequency. We budget integration work separately because it's inherently unpredictable. The sports data API you want might have great docs. It might also go down every other weekend.
5. Design complexity
A clean, functional UI with standard patterns is one thing. Custom animations, interactive data visualizations, 3D elements, and complex responsive behaviors are another. The animation system we built for WSV took significant design and engineering time -- GSAP scroll-triggered sequences, device mockups, and fluid transitions. Beautiful design is not free.
6. Content management
If non-developers need to update content, you need a CMS. That could be a headless CMS like Sanity or Contentful (faster to set up, ongoing subscription cost) or a custom admin dashboard (more development time, no subscription). Footy Access needed a custom content publishing workflow because off-the-shelf CMS options couldn't handle their specific editorial process.
7. Data complexity
Simple CRUD operations are cheap. Complex data relationships, search with filtering, aggregations, reporting dashboards, and data migrations are expensive. The more interconnected your data model, the more time goes into getting it right -- and the more expensive it is to change later.
When custom beats off-the-shelf
Not everyone needs custom software. If your business runs on standard workflows -- basic e-commerce, simple content publishing, appointment booking -- there are excellent off-the-shelf tools that will serve you well at a fraction of the cost.
Custom makes sense when:
-
Your workflow is your competitive advantage. If the way you do business is fundamentally different from your competitors, generic software will force you into generic patterns. NXTGEN Media's talent network doesn't work like a standard social media platform or a standard recruiting tool -- it's a hybrid that required custom architecture.
-
You need to own the experience end-to-end. When your software IS your product (not just a tool that supports your product), you can't afford to be constrained by someone else's platform decisions. Edge Athlete OS needed a specific coaching interaction model that no off-the-shelf fitness app could provide.
-
You've outgrown your current tools. We've had multiple clients come to us after hitting the ceiling of WordPress, Squarespace, or Wix. Their business grew, their needs got more complex, and the tool that got them started became the thing holding them back.
-
Integration requirements are specific. When you need three or four systems to talk to each other in non-standard ways, custom middleware or a custom platform becomes cheaper than fighting with Zapier automations and webhook chains.
How we price at Aetheris
We've tried hourly billing, and we've tried pure fixed-price. Neither works well in isolation. Here's what we've settled on:
Fixed-scope, milestone-based pricing
Every project starts with a paid discovery phase. We spend one to two weeks understanding the business, mapping the technical requirements, and producing a detailed scope document. That document becomes the contract -- every feature, every screen, every integration, explicitly listed.
The total project cost is broken into milestones, each tied to a deliverable. You pay when the milestone is delivered and approved. This gives both sides clarity: you know exactly what you're paying for, and we know exactly what we're building.
What's included (and what isn't)
Our quotes include design, development, testing, deployment, and a 30-day post-launch support window. They don't include ongoing maintenance, hosting costs, or feature additions beyond the original scope. We're transparent about this upfront because surprise invoices are the fastest way to destroy a client relationship.
Change orders are normal
Scope changes happen on every project. We don't penalize clients for learning things during the build. But we do document changes formally, re-price them, and get approval before building. This keeps the project honest and prevents the slow scope creep that kills timelines and budgets.
The question to actually ask
"How much does an app cost?" is the wrong question. The right question is: "What's the minimum version of this product that proves the business model, and what does that cost?"
Start there. Build that. Learn from real users. Then invest in the next version with actual data instead of assumptions.
We've seen $15K MVPs turn into $200K platforms over two years. We've also seen $100K builds that should have been $25K MVPs first. The clients who do well are the ones who treat the first version as a learning tool, not a finished product.
If you're trying to figure out where your project falls, we're happy to talk through it. No pitch, no pressure -- just an honest assessment of what you're looking at. That's how every good project starts.