The cost of a custom website or app is driven mainly by scope, integrations with other systems, how much content and design work is needed, the platforms you target, and the quality bar for security, performance and testing. Ongoing hosting and support matter as much as the build. A short discovery phase is the most reliable way to turn these into a firm estimate.
“How much does a website cost?” is a bit like “how much does a house cost?”. The honest answer is that it depends, which is unhelpful unless you know what it depends on. Two projects that sound identical in a one-line brief can differ several times in effort once the details come out.
Here are the factors that actually move the number, roughly in order of impact, and what you can do about each one.
1. How much does scope really matter?
Scope is the biggest driver by far. Every page type, user role, feature and edge case is work to design, build and test. A marketing site with ten pages is a different project from a site with customer accounts, a booking system and an admin panel, even if both “look simple”.
The trap is that features rarely arrive alone. A login brings password resets, email verification, account settings and security reviews with it. A “simple” booking feature brings availability rules, time zones, cancellations, reminders and payments.
2. Why do integrations add so much?
Connecting to other systems, such as your CRM, ERP, payment provider, accounting software or an internal database, is often where the hidden effort lives. A well-documented modern API can take a day or two to integrate. An older system with no API, inconsistent data or strict security rules can take weeks.
Before estimating, it helps to know which systems are involved, whether they have APIs, who owns them and whether there is a test environment to build against.
3. Is design and content a big factor?
A site built on an existing design system or a strong template needs far less design time than a fully bespoke visual identity with custom illustration, animation and many unique page layouts. Content is the other half: someone has to write, photograph, translate and organise it. Content that is not ready is one of the most common reasons projects run late.
4. Web, mobile or both?
Each platform you target adds work. A responsive website works everywhere, but a mobile app adds app store releases, device testing and push notifications. Building for iOS and Android with one shared codebase keeps this manageable, but it is still more than a website alone. A desktop app adds installers and updates on top.
5. What does quality cost?
Two builds with the same features can still differ a lot in effort because of the quality bar:
- Security and compliance. Handling payments, health data or personal data properly takes deliberate work, reviews and sometimes audits.
- Performance. A site that loads fast on a mid-range phone over a slow connection takes more care than one that only works well on a laptop.
- Testing. Automated tests cost time up front and save far more later, especially on anything you will keep changing.
- Accessibility. Building it in from the start is cheap; retrofitting it later is not.
6. What happens after launch?
The build is only part of the cost. Hosting, monitoring, backups, security updates, small changes and support continue for as long as the product runs. A useful rule of thumb is to plan a yearly budget for running and improving the product, not just a one-off build cost. The right hosting choice, whether managed cloud or a well-run VPS, can make a real difference here.
How do agencies usually charge?
There are three common models, and each suits a different situation:
- Fixed price. Works when the scope is clear and unlikely to change. Expect the estimate to include a buffer for risk.
- Time and materials. You pay for the time spent. Best when the scope will evolve, and it gives you the most flexibility.
- A monthly retainer. A dedicated team or a set number of hours each month, suited to products that keep growing after launch.
How can you keep the cost under control?
- Start with a short, paid discovery phase to agree scope, risks and priorities before committing to the full build.
- Launch a smaller first version and improve it with real user feedback.
- Have one decision-maker on your side, and get content ready early.
- Reuse proven building blocks, such as a template, a design system or an existing integration, wherever they fit.
- Ask how the estimate was built. A good partner can explain which assumptions it rests on.
A clear brief is the best way to get an accurate number. If you know your must-haves, the systems involved and your timeline, a good team can usually give you a realistic range quickly and a firm estimate after discovery.
Questions
Why do quotes for the same project vary so much?
Different teams make different assumptions about scope, quality and risk, and some leave out things like testing, content or post-launch support. Ask each one what is included and what they assumed; the gaps usually explain the difference.
Is a template cheaper than a custom design?
Usually, yes. Starting from a well-built template or design system can save a lot of design and front-end time, and it can still be adapted to your brand. Fully bespoke design makes sense when the visual experience is central to the product.
What is a discovery phase, and is it worth paying for?
Discovery is a short phase, often one to two weeks, where the team maps requirements, systems and risks and produces a plan and estimate. It is worth it for anything beyond a simple site, because it replaces guesswork with a plan you can budget against.