You describe your app idea to three developers and get three quotes: one small, one medium, one that makes you close the laptop. They all heard the same idea. So which one is right?
Possibly all of them. They just imagined three different apps.
I've been on the quoting side of this conversation many times, and the price almost never comes from the idea itself. It comes from a list of details most people haven't decided yet. This post walks through those details, gives rough time ranges for common types of apps, and shows you how to get a quote you can actually compare.
Why quotes are so far apart
"An app like Uber for dog walkers" could mean:
- A simple website where owners request a walk and you assign walkers by hand.
- A mobile app for owners, a separate app for walkers, live GPS tracking, in-app payments with payouts to walkers, ratings, and an admin dashboard.
Both are honest readings of the same sentence. The second is easily ten times the work of the first.
Most of the gap between quotes is this: each developer filled in the missing details differently. The fix isn't to find the "right" price. It's to remove the guesswork before you ask.
What actually drives the cost
1. Who uses it
Every type of user is close to a separate product. A marketplace has buyers, sellers and you, the admin. Each needs its own screens, its own permissions and its own flows. Going from one user type to three roughly triples the surface area.
2. How many things they can do
Count the core actions, not the screens. "Sign up, create a listing, search listings, message a seller, pay" is five actions, and each one comes with its own edge cases. What happens if a payment fails halfway? If a listing is edited while someone is paying for it?
3. Integrations
Payments, maps, email, SMS, calendars, accounting software, shipping providers. Each integration is a small project with its own testing, failure handling and account setup. Payments with payouts to other users (like a marketplace) are much more involved than simply charging a card.
4. The admin panel nobody mentions
Someone has to approve users, issue refunds, edit content and answer "why did this happen?" If it isn't in the plan, you'll be doing it directly in the database, or paying for it later. Ask whether a quote includes it.
5. Platforms
Web only, iOS, Android, or all three? Cross-platform tools like React Native and Flutter let one codebase cover both phone platforms, which saves a lot compared to two separate native apps. It's not free, though: app store reviews, push notifications and device testing still add time.
6. Real-time and AI features
Live chat, live location, collaborative editing and AI features (chatbots, document search, generated content) all add complexity beyond the basic screens. AI features also add a running cost for every request, which belongs in your budget.
7. Design
Using a well-made component library with your colours and logo is much faster than a fully custom design. Both are valid. Just know which one you're paying for.
8. Rules you have to follow
Health data, financial data, children's data, or customers in regions with strict privacy laws all add requirements around storage, access and auditing. Mention these up front, because they change how things are built, not just what's built.
Rough time ranges for common projects
Time is a better way to compare than price, because rates vary hugely by country and experience. These are rough ranges for an experienced developer or small team building a solid first version:
- Marketing site or landing page with a contact form: about 1 to 2 weeks.
- Internal tool or dashboard over existing data: about 3 to 6 weeks.
- MVP web app with accounts, one core workflow, payments and a basic admin panel: about 6 to 10 weeks.
- Mobile app (iOS and Android) with a backend: about 8 to 14 weeks.
- Two-sided marketplace with payouts, messaging and reviews: about 3 to 6 months.
Multiply by the rate you're quoted to get a ballpark. For example, at a hypothetical $50 per hour, 8 weeks of full-time work is around $16,000. At $100 per hour it's double that. Same app, same weeks.
If a quote is far below these ranges, ask what's been left out. If it's far above, ask what they assumed you need that you might not.
Costs that come after launch
The build is not the whole bill. Plan for:
- Hosting and services: servers, database, file storage, email sending, error tracking. Small at first, growing with users.
- App store accounts: Apple charges $99 per year for a developer account, Google charges a one-time $25 fee.
- Third-party fees: payment processing takes a percentage of each transaction, maps and SMS charge per use, AI models charge per request.
- Maintenance: operating systems update, libraries need security patches, and APIs you depend on change. A common rule of thumb is to budget around 15 to 20 percent of the original build cost per year just to keep things healthy, before adding new features.
Did AI tools make apps cheaper?
Somewhat. In 2026, AI coding tools make the routine parts much faster: boilerplate, standard screens, simple CRUD features, first drafts of tests. A good developer using them well ships faster than they did two years ago.
What hasn't changed is everything that requires judgement. Deciding what to build. Designing the data model. Getting security and permissions right. Handling payments properly. Testing with real users. These are where most of the cost and most of the risk lives, and AI doesn't remove them. If anything, apps built quickly with AI need more attention here, not less. I wrote about that in why AI-built apps break with real users.
How to get a quote you can compare
Write one page before you talk to anyone. It doesn't need to be technical:
- Who is it for? List each type of user in one line each.
- What are the three to five things they must be able to do in the first version?
- Which platforms? Web, iOS, Android.
- What does it connect to? Payments, existing systems, other services.
- What already exists? Designs, a website, a database, a half-built app.
- When do you need it, and why then? A real deadline changes the plan.
- What's your budget range? This isn't a trap. It lets a developer tell you what's realistic, or suggest a smaller first version that fits.
Send the same page to everyone you ask. Now the quotes are answering the same question.
Fixed price, hourly, or phased?
- Fixed price works when the scope is clear and unlikely to change. You know the cost up front, but every change becomes a negotiation.
- Hourly or weekly works when you're still figuring things out. You pay for what's done and can change direction, but you need to trust the developer and watch progress.
- Phased is what I usually recommend for new products: a short paid discovery to pin down scope, then a fixed price for the first version, then ongoing work for what you learn after launch.
Red flags in a quote
- No questions asked before the number. Nobody can price an app from one sentence.
- No mention of an admin panel, testing or deployment.
- Everything is "included" with no list of what that means.
- No plan for what happens after launch.
- You don't own the code or the accounts it runs on.
The honest answer
How much does an app cost? Whatever the clearest version of your first release costs, and you get to decide what that is. The fastest way to a fair price is to cut the first version down to the few things that prove your idea works, build that well, and grow from there.
If you have an idea and want a realistic estimate, send me your one-pager. I'll tell you what I'd build first, roughly how long it would take, and what I'd leave for later. You can also see the kinds of products I've built.
