Website or mobile app: which does your business actually need?
A decision framework for choosing between a website, a web app, a mobile app — or the increasingly common right answer of starting with one.
This question is usually asked backwards. The useful version is not "website or app" but "how often will one person use this, and what do they need it to do?" The answer to that determines the format almost entirely.
The frequency test
Mobile apps live on a home screen. That placement is earned by repeat use, and it is the single strongest predictor of whether an app makes sense.
- Used once or twice ever — a website. Nobody installs an app to make one booking.
- Used occasionally, a few times a year — a website, ideally installable as a progressive web app.
- Used weekly or daily — an app becomes defensible.
- Used daily as part of someone's job — an app is usually the right answer.
The capability test
Some things browsers still do poorly or not at all. If your product genuinely depends on one of these, that settles it.
- Reliable push notifications, particularly on iOS
- Continuous background location
- Heavy offline use with local data
- Bluetooth, NFC or camera-intensive processing
- Biometric authentication tied to the device
If none of these apply — and for most service businesses, marketplaces and content products, none do — the web can deliver the same experience at lower cost and with no install friction.
The distribution test
This is the one that quietly decides most projects. A website is one link away from any advertisement, search result or message. An app requires a store visit, an install and an account before it delivers any value at all. That gap is enormous, and for a business acquiring customers through ads or search it usually outweighs everything else.
If you are still proving that people want what you're offering, choose the format with the least friction between seeing it and using it.
The answer that suits most businesses
Start with the web. Build a fast, responsive web application that works properly on a phone, and instrument it so you can see how people actually use it. If a cohort emerges that uses it several times a week, you now have both the justification for an app and the behavioural data to design it well.
When to build both
Building both is right when the two serve genuinely different jobs: a marketing site that acquires customers, and an app that serves them after they convert. That is a deliberate split with distinct audiences. Building both to deliver the same experience to the same people is simply paying twice.