Applications

Does Every Business Need an App?

A native app isn't always the answer. Website, PWA and native app compared, and the criteria I decide by: frequency of use, notifications, offline work, camera and the cost of maintenance.

Comparison table: website, PWA and native app by capability
Three options that differ considerably in cost and capability.

The question of whether a business needs its own app comes up sooner than you'd expect. Sometimes it comes from a client who has seen a competitor's app. Sometimes from me, when I wonder on one of my own projects whether users would find things easier with an icon on their phone.

Since I also plan and develop apps, my interest would be obvious: of course you need an app. But that's not how I think about it. In many cases a native app is an additional cost, an additional thing to maintain and an additional obstacle for a user who has to install something before they can even see whether it's useful.

So I'd like to describe how I think about this question when I put it to myself.

Three options, not one

When someone says "app", they usually have an icon in an app store in mind. In reality I'm choosing between three options that differ considerably in cost and capability.

A good responsive website works in every browser, on every device, without installation. Search engines find it, it can be shared with a link, and I maintain one version. For most presentations, orders and information, that's enough.

A progressive web app (PWA) is a website the user can add to their home screen, which can work offline and which, on modern devices, can do some of the things we attribute to apps. One codebase, one deployment, no store.

A native app for iOS and Android has the deepest access to the device and a presence in the store. It also has two codebases, or a framework bridging them, two stores with their own rules and reviews, and updates the user has to install.

When an app really makes sense

The question I ask first isn't what the business wants but what the user will do. An app makes sense when they'll do something often and when the device can help them in a way a browser can't.

Frequent return is the first condition. If the user uses something every day or every week, an icon on the screen has value. If they use it twice a year, installation is an obstacle, not an advantage.

A personal profile and data the user carries with them: orders, balances, documents. Push notifications when they're genuinely useful and not just marketing. Working offline when the user will be somewhere without a connection. Camera, location and sensors when they're part of the task and not decoration. Native integrations needed for payments, access or hardware.

If I recognise two or three of these conditions, an app is a justified consideration. If none, it's probably just a more expensive version of a website.

Where a PWA covers more than people think

Quite a few things on the list above can be done by a PWA today. An icon on the home screen, launching without the address bar, working offline, storing data on the device. Push notifications have worked on Android for a long time; on the iPhone, Apple has supported them since iOS 16.4, but only for web apps the user has added to the Home Screen and that have a properly configured manifest.

That condition matters, because marketing descriptions like to leave it out. A user on an iPhone first has to add the site to the Home Screen; only then can the app ask for permission. It's not the same as installing from a store, but for many users it's simply enough.

Access to the camera or location is possible in the browser with the user's permission. Some things remain for native apps: deeper system integrations, background work, presence in the store, some payment methods. If a project needs exactly those, a PWA won't do.

With a PWA I still check on real devices, not in documentation. The differences between browsers and system versions are big enough that "supported" in a table can mean "supported with caveats" in practice. A few minutes with a phone in hand say more than an hour of reading about support. I compared the two point by point (installation, updates, notifications, device access, maintenance cost) in PWA or Native App.

The cost we don't see at the start

When deciding, the conversation likes to stop at the cost of building. What worries me more is the cost that comes afterwards.

A native app has to keep up with new system versions, changes in store rules, new devices. Every change goes through a review that isn't always fast. Users have different versions, and old bugs live longer. Support answers questions that a website doesn't raise.

A website and a PWA have one version that is always current. I fix a bug once and it's fixed for everyone. For a smaller business without a maintenance team, that's the argument that often decides it.

Installation is an obstacle, not a milestone

In planning we tend to forget that installing an app is an action that demands something of the user. They have to find it, trust it, wait for the download, perhaps create an account. Each of those steps loses some people.

A website has no such threshold. Link, click, use. For someone who is only getting to know the company, that's the only path that makes sense. Only once the user knows they'll be coming back does installation become sensible. That's why I see an app as something that comes after the website is already doing its job, not instead of it.

How I decide on my own projects

On my own projects I ask myself a few questions, which I'll repeat here in the same form I put them to myself.

How often will the user use this? What will they do that they can't do in a browser? Will they need data offline? Are notifications part of the service, or just a way of reminding them we exist? Who will maintain the app in two years? Would we achieve more with the same money if we put it into the website?

If the answers point to a website, I start there. If they point to a PWA, I add what's missing and check it on real devices. A native app remains for cases where the need is clear and where the cost of maintenance is covered by the value to the user.

That order seems fair to me, to the business and to the user. The business doesn't pay for something it doesn't need, and the user doesn't install something they could have opened in a browser.

An app as a sign of seriousness

One more thing I hear often: an app is supposed to show that a company is serious. I understand the feeling; I don't agree with it.

Seriousness is shown by a fast site that answers questions and by a service that keeps its promises. An app that few people download and most forget doesn't show seriousness; it shows that an expensive decision was made somewhere without the user in mind. An empty app in the store with two ratings from three years ago doesn't help a company.

Three hypothetical cases

To make the criteria more tangible, take three invented companies.

A restaurant wants an app for reservations and viewing the menu. A guest will use it a few times a year, wants to see the menu before deciding, and makes a reservation in a minute. A fast website with a tap-to-call button and a simple form does all of that; installing an app would be an obstacle with no benefit for the guest. Answer: a website.

A service company has field workers who fill in work orders every day, photograph the completed work and also work where there's no signal. Use is daily, the camera is part of the task, offline operation is essential, the users are known and can be helped with installation. Answer: an app, and I'd first check whether a PWA with local storage and camera access is enough before going to two native codebases.

A tool for small business owners, used every day to enter data, with a profile, deadline notifications and work on the phone while out and about. Here the question is serious and the answer depends on which devices the users have and what the notifications really mean. A PWA covers most of it; a native app comes into consideration if users genuinely expect an icon from the store and if the cost of maintenance is justified by what the app adds.

In all three cases the decision follows from what the user will do and how often. Nowhere is the starting point what the competition has.

The cost questions I ask before building

With an app I ask a few more things that a website doesn't require. Who will publish updates, and how quickly, when a store changes its rules? What happens when a user doesn't update and uses a version from six months ago? Who will answer ratings and questions in the store? Will we maintain two platforms or one? How many users do we need for this to pay off, and do we have them?

These are dull questions, but the answers decide whether the app will still be alive in two years. A website doesn't raise them, and that is often the best argument for it.

A decision based on the user

The question in the title has a short answer for me: no. Not every business needs an app. Every business needs a path by which a user finds it, understands it and uses it, on the device they have in their hand.

For most, that path is a good website. For some, it's a PWA that adds what's missing. For a minority, it's a native app, and with those you can tell, because users do something every day. In my experience, a decision based on what the user will actually do tends to hold up better than one based mainly on what the competition has.

Photo of the author

I build web solutions and digital products. At GE-KO I work on web design, development and SEO; with Bizko I look for simpler ways through everyday business. I am interested in how technology saves people work, and in how AI is changing the web.

One step further.