There's no honest way to answer "how much does a web app cost" in one number, and any agency that fires back a figure before understanding what you're building is guessing. But you didn't come here for a shrug, so let's be specific. In 2026, most custom web apps built in the UK land somewhere between £8,000 and £120,000. That's a huge spread, and the rest of this article is really about where your project falls inside it — and why.
The three bands we see most often
Strip away the jargon and nearly every quote falls into one of three brackets. These are real-world ranges for a UK team of experienced developers, not offshore hourly rates or no-code tinkering.
| Type of app | Typical UK cost | Rough timeline |
|---|---|---|
| Simple app or MVP (one core job, one user type) | £8k – £25k | 4–8 weeks |
| Growing product (accounts, dashboard, payments, admin) | £25k – £60k | 2–4 months |
| Complex platform (multiple roles, integrations, real-time) | £60k – £120k+ | 4–9 months |
If your first reaction is "why is the top band fifteen times the bottom one?", that's the right question — and the answer is almost never "better code". It's scope. A booking form that emails you is a different animal to a two-sided marketplace with live availability, Stripe payouts and an ops dashboard, even though a founder might describe both as "an app".
What actually drives the number
When we scope a project, five things move the price far more than anything else. Understanding them lets you steer the cost instead of just receiving it.
1. How many types of user you have
Every distinct role — customer, admin, driver, seller, moderator — is effectively its own mini-application with its own screens, permissions and edge cases. Going from one user type to two rarely doubles the cost, but it's the single biggest lever most people underestimate.
2. Whether money changes hands
Taking a card payment is straightforward. Handling other people's money — subscriptions, refunds, marketplace payouts, VAT, failed-payment logic — is where hours quietly pile up, because the software has to be right every single time.
3. Integrations with other systems
Connecting to a CRM, an accounting tool, a shipping provider or a legacy database means living with their quirks, rate limits and downtime. Two or three integrations can add more work than a whole extra feature you build yourself.
4. Real-time and data-heavy features
Live chat, notifications, maps that update as things move, dashboards crunching lots of records — anything that has to happen "instantly" costs more than a page that loads when asked. It's often worth it, but you should know you're paying for it.
5. Design polish and accessibility
A functional interface and a genuinely good one are different budgets. For a consumer product where the experience is the product, that gap is worth closing. For an internal tool five people use, it usually isn't.
If you can describe your app in one sentence without the word "and", you're probably in the bottom band. Every "and it also…" tends to move you up a bracket. Writing that sentence down before you ask for quotes is the cheapest planning you'll ever do.
Why the same brief gets a £15k and a £45k quote
This trips up almost everyone, so it's worth naming. When two agencies quote the same brief very differently, it's usually one of these:
- They're pricing different things. One included the admin panel, testing and a month of post-launch fixes; the other quoted the happy path only. The cheaper number often isn't cheaper — it's smaller.
- Different teams, different rates. A senior UK developer who ships the thing once, correctly, can work out cheaper than a low day-rate team that needs three attempts.
- One of them padded for risk because the brief was vague. Ironically, a clearer brief gets you tighter, lower quotes.
The fix isn't to chase the lowest figure. It's to make every quote describe the same scope, then compare like with like. A good development partner will happily break their number down; be wary of anyone who won't.
How to spend less without cutting corners
Cheaper and worse aren't the same thing. Here's where you can genuinely save money that we'd tell our own clients:
- Build the smallest version that's actually useful first. Ship the one feature that proves people want it, learn from real usage, then invest in the rest. This is the whole idea behind an MVP — and it's the difference between spending £15k to learn something and £60k to guess.
- Use proven building blocks. Authentication, payments, hosting and databases don't need reinventing. A team that reaches for battle-tested tools (rather than custom-building everything) spends your budget on what makes your product different.
- Cut features, not quality. Ten features done to 70% will frustrate users more than four done properly. Ruthlessly postpone; rarely compromise.
What if you don't have the budget yet?
Plenty of the best products we've seen started with a strong idea and a founder who couldn't write £40k off the bat. That's a real position, not a dead end. It's exactly why we run an investment route: for ideas we believe in, we can fund and build the product ourselves — up to £50,000 — and back you as a long-term partner rather than just an invoice. If capital is the thing standing between you and a launched product, that's worth a conversation before you go hunting for the cheapest possible quote.
The honest bottom line
A web app in the UK in 2026 costs what its scope demands — realistically £8k for something small and sharp, £25k–£60k for a proper product, and into six figures for a genuine platform. The number you can control most isn't the day rate; it's how much you ask the software to do on day one. Decide that well, and you'll spend less and ship sooner.