Does my business need an app or a website?

An app is not a channel a customer happens onto: it is a commitment the customer makes, and every uninstall is churn that never tells you.

The question arrives in almost every planning conversation: does my business need an app or a website? The owner asking it usually has a budget for one, and has been told by someone that the app is the modern version of the website, so a serious business eventually gets both.

That framing is the market's false version, and it quietly misdirects money. For most service businesses the website, done well on a phone, is the product: it is where search lands people, where offers are read, where contact happens, and it asks nothing of the customer before showing its worth.

The honest mechanism is about the shape of each thing. A website is a channel a customer happens onto: found, opened, used, and gone, without installing anything, which is close to what buying a service actually looks like. An app is not a channel. It is a commitment the customer makes, an install on a personal device, and uninstalls are churn that never sends a message.

So the app question is really the retention question. An app earns its place only when there is a repeated, personal, two-sided job: accounts that matter, saved state a customer returns to, moments worth a notification, work that survives a dropped connection. If you cannot name the job a customer would open weekly on its own, the app becomes a brochure nobody opens, stored on the phone between the torch and the compass.

The failure versions are expensive. The app that duplicates the site and gets uninstalled inside a month. The budget split so that both the site and the app end up mediocre, when a strong site alone would have carried the year. The store listing treated as distribution, as if presence on a store shelf were the same thing as discovery, when discovery does not come from the store at all.

The practice is to write the decision down before spending. Write down the repeated job the app would do, in one sentence, tied to a concrete customer action. Be honest about whether customers would open it weekly without being reminded. If the answer holds, build the website first and prove the behavior there; the evidence either dissolves the app question or gives the app a real brief to earn its build.

Guidance, not guarantees: we build web products for a living, so weigh our bias in the balance, but the reasoning above is not about defending our own work. Name the weekly job first. If it exists, an app conversation is worth having and we will help you scope it honestly. If it does not, the best money you can spend is on a website that respects the phone.