Does your business actually need an app?

We turn down app projects. Not often, but regularly enough that it is worth explaining the reasoning, because an unused app is one of the more expensive mistakes a business can make — and the cost does not stop at the build.

The real cost of an app

The build is the smaller half. An app also carries:

• Mandatory maintenance. Operating systems update annually. Libraries deprecate. Store policies change. Your app needs work every year whether or not you change a single feature — otherwise it eventually stops functioning.
• An install barrier. Every user must decide to download, find storage, and grant permissions before seeing any value. A website asks for none of that.
• Release friction. Fixes go through review. A critical bug takes hours or days to reach users, not minutes.
• Two platforms. Reaching your whole audience generally means iOS and Android, with testing across a device matrix.

An app has to earn all of that back.

The four-question test

1. Will people use it more than weekly?

Apps live on the home screen, which is only valuable if it is visited. If your customers interact with you monthly — booking an appointment, ordering occasionally, checking something rarely — the icon becomes clutter and gets deleted. Frequency is the strongest single predictor of whether an app survives.

2. Does it need to work offline?

Field service, logistics, inspections, sales in low-connectivity areas, anything used in a basement or a warehouse. Genuine offline capability is difficult on the web and native on mobile. This is one of the strongest cases there is.

3. Does it need real device capability?

Continuous background location, reliable push notifications, hardware camera control, biometric authentication, Bluetooth peripherals. Not "we want to send notifications" — the web can do that. Something the browser genuinely cannot.

4. Is it a tool for your own team?

Internal apps skip the hardest problem — acquisition. Your team will install it because it is their job. Operational apps that remove daily friction from a field or floor workforce often return their cost quickly, and they are consistently the most successful app projects we see.

One clear yes justifies the conversation. Zero means a fast mobile site will almost certainly serve you better.

"But our competitor has an app"

Check whether it is any good. Look at the review count, the recency of the last update, and what the one-star reviews say. A significant proportion of business apps are abandoned within two years — shipped, announced, then quietly unmaintained.

Competitor presence is not evidence of competitor success.

If the answer is yes

Then the next decisions matter: cross-platform unless performance or hardware access demands native; onboarding and first-session experience treated as core product work rather than polish; a notification strategy that respects the user, because the fastest route to deletion is being noisy; and analytics from day one so you know whether anyone came back after week one.

Retention is the only number that matters. Installs are a vanity metric that flatter a launch and tell you nothing.