A straight answer, as far as one exists
How much does custom software cost in NZ?
Most studios won't publish numbers. We'd rather give you an honest range and explain what moves it, so you can work out whether your idea and your budget are in the same room before anyone gets on a call.
The short answer
For a small, senior New Zealand studio like ours, most projects land in one of three bands. These are honest ballparks, not quotes - every project is different, and the drivers below explain why.
Small
$8k – $30k NZDA focused prototype or a single, well-defined feature
2–5 weeks
A working prototype of a booking tool, a customer portal MVP, or one substantial feature added to an existing system.
Medium
$30k – $90k NZDA complete product or app with several features and a few integrations
6–12 weeks
A customer-facing web app with accounts, payments and an admin area, or a mobile app connected to your existing systems.
Large
$90k+ NZDA complex, multi-part system with multiple integrations or scale requirements
3+ months
A platform with several user types, real-time features, third-party integrations, or serious compliance requirements.
What actually drives the cost
Two ideas that sound the same size in a sentence can be five times apart in cost. These are the things that move the number most, roughly in order.
How many kinds of user it has
An app for one type of user is one app. Add customers, staff and admins, each with their own screens and permissions, and you're building several connected apps.
Integrations
Every external system you connect to - Xero, a payment provider, a legacy database - adds work, and the messiness of the other system matters more than your own.
Accounts, payments and permissions
Login, billing and role-based access are largely solved problems with modern tooling, but they still add scope. A tool without accounts is dramatically cheaper than one with them.
Real-time, offline and sync
"It should update live" and "it should work without internet" sound like small asks. They're two of the biggest cost multipliers in software.
How settled the idea is
A clear, decided scope is cheap to build. A scope that changes weekly is expensive no matter who builds it. This is the driver you have the most control over.
Platforms and publishing
Web only is the cheapest start. Native mobile apps and app-store publishing add real work - often deferrable until the idea is proven.
How to keep the cost down
The biggest saving isn't negotiating the rate - it's cutting scope before the build starts. Start web-only even if you want an app eventually. Skip login until you have users worth logging in. Use managed services instead of custom infrastructure. Build one workflow well instead of six workflows badly.
The other half of keeping cost down is scoping well before you ask anyone for a number — we've written a separate guide on how to scope a software project so the quotes you get actually mean something.
This is exactly what our rapid prototyping engagements are for: the smallest build that can test the idea, so the bigger spend only happens once the idea has earned it. If the prototype proves out, taking it to production is a decision you make with evidence, not hope.
When you shouldn't build custom software
Sometimes the right answer is a spreadsheet, an off-the-shelf product, or a no-code tool - and it's cheaper for everyone if that's said early. If your workflow is standard, your volumes are small, and you don't need to own the product, custom software is probably premature. We'll tell you if that's what we see.
Common questions
How much does it cost to build an app in New Zealand?
As a rough guide: a focused prototype typically lands between $8k and $30k NZD, a complete product with several features between $30k and $90k, and complex multi-part systems upwards of $90k. The honest answer is that it depends almost entirely on scope - which is why we scope before we quote.
Why won't developers give a fixed price up front?
A fixed price on a vague scope is a guess with a margin baked in - you either overpay for the padding or the project fights over every change. A short scoping exercise first means the number you get is grounded in the actual work.
What's the cheapest way to get a software idea off the ground?
Build the smallest thing that can test the idea with real users: one platform, managed infrastructure, and no features that aren't needed for the test. Skip accounts, native apps and integrations until the idea has earned them. That's what our rapid prototyping engagements are.
Is it cheaper to use a no-code tool instead?
Often, yes - and we'll tell you when it is. No-code is a great way to test a workflow. Custom software earns its cost when you hit the walls: integrations no-code can't reach, performance, ownership of your data, or a product you intend to sell.
What happens to the code if we stop working together?
It's yours. You own the code and the infrastructure accounts from day one, so you can take it to another developer or an in-house team at any point.
Want a number for your actual idea?
Describe your project and get a rough sense of size and timeline in about three minutes - no call, no obligation.
Get a rough estimate