Having a web app or app built: when custom software pays for itself

Almost every app enquiry starts with the same sentence: we want an app. And an app is almost never the right first answer. A process is stuck, and the real question is which form gets it moving again fastest.

A woman in a blue suit measures a giant, empty phone screen with a yellow tape measure, against a violet background.

First, the question behind the question

Before you talk about technology, you want to know which behaviour you are trying to change. In practice, enquiries almost always come back to one of three things: customers phone to ask how their order or request is going, the team keeps the truth in a shared spreadsheet, or someone copies data from one system into another every morning.

Those are three different problems with three different solutions. Status questions are often solved with a simple page behind a link. A spreadsheet that has become the truth calls for a real application. Manual retyping often needs no interface at all, just an integration. Examples of how that plays out are in our cases.

Four forms, four different bills

A website with smart forms is public and delivers requests in a structured way. A customer portal is a logged-in part of your site with status updates and documents. A web app is a full application in the browser, on mobile too. A native app is one the user installs from the App Store or Play Store. Maintenance rises in that same order, from low to high. The table below shows what mainly pushes up the cost of each form.

Form What pushes up the cost When it fits
Website with smart forms The forms and the link to your back office The process starts with a request
Customer portal Log-in, permissions and the link to your systems Customers want to see the status themselves
Custom web app Every role, exception and integration you want from day one Your process is your product
Native app Two platforms, app store requirements and maintenance with every OS update Push notifications, camera, offline use or hardware

Note that the jump from portal to app is not only in the build cost, but above all in the tail. Every other form in the table you maintain at a website rhythm. An app you maintain at a product rhythm.

What native really involves

Publishing a native app takes an Apple Developer Program membership at $99 a year and a one-off $25 registration with Google Play. That is the cheap part. Every release goes through review, each platform has its own guidelines, and your users actually have to update. Even with a cross-platform framework, you are testing on two ecosystems.

A native app pays for itself if you need push notifications, the camera, background location, offline use or hardware, or if users open it every day. If you need none of that, an app is often a more expensive wrapper around a website you have to build anyway.

Most businesses do not want an app. They want customers to be able to do something themselves, without picking up the phone.

Where the money goes in custom work

On a custom platform, the budget rarely goes on screens. It goes on decisions and on integrations. Roughly four items: getting the scope sharp, designing the flows and the data structure, the build, and the integrations with what you already use. That last item is the biggest unknown, because it depends on systems you do not own.

Then plan for upkeep. A platform nobody looks after is a security risk within two years and a millstone within three. Set aside a structural budget and attention for it, or do not build it.

Four signs custom work will pay for itself

  • Hours go every week on retyping data between systems.
  • Only one person truly understands the spreadsheet that holds everything together.
  • Customers phone for information that is already written down somewhere.
  • Adding a customer or a location means more work, not economies of scale.

A simple sum helps keep the conversation honest. Count the hours of manual work per week, multiply by an internal hourly rate and by the number of working weeks in a year. Set that against the build cost plus upkeep. If the payback period is under two years, it is worth the conversation. If it is five years, it is probably a process you should simplify first.

Start with the smallest version that truly solves one process

The most expensive way to build a platform is all at once. The cheapest is a first version that completes exactly one process, end to end, for real users, in weeks rather than quarters. After that, you build on what you have learnt, not on what you dreamt up in a workshop.

With us, focused sites usually land in six to eight weeks; platforms and stores typically in three to four months. What goes into having a web application or app built, from scoping to launch, is on the service page; if you are still unsure about the form, it starts with digital strategy. What a project costs depends on the form you choose and the number of integrations; the table above shows what drives the cost of each form. If you sell products rather than time or expertise, the comparison is a different one: read having an online shop built. A concrete example of a platform question with a clear return is a booking system on your own site.

Frequently asked questions

Web app or native app?

Go native if you need push notifications, the camera, background location, offline use or hardware, or if people use your product every day. In every other case, a web app goes live sooner, costs less to maintain and can be shared straight away with a link. You can always build native later, on a platform that has already proved itself.

What does a first version cost?

That depends mostly on the number of integrations and user roles; the table above shows what pushes up the cost of each form. The biggest cost is never the number of screens, but the number of exceptions you want to support from day one.

How long before we can use something?

With a tightly scoped first version, real users are typically working in the system within three to four months. If it takes longer, the problem is usually the scope, not the pace of the build.

Get started

Have something in mind? Let’s make it.

Book a free 30-minute intro call. We’ll show you how we would approach your idea.