The honest default is no
An app is a channel with a toll gate on it. To reach a customer you ask them to open a store, search your name, download a file, and sometimes make an account. A mobile website asks for a tap. Weigh that gap before anything else.
For most local service businesses and most small online shops, a mobile website does the work. It lists services, prices and hours. It takes a booking or a quote request. It accepts payment. Search engines can read it. You can change a price at lunch and have it live by one.
If you cannot say in one sentence what the app does that a good mobile site cannot, you already have your answer.
Four conditions that justify building one
Frequency is the first. An app makes sense when a customer opens it several times a week without prompting. A class booking, a route-based delivery, a trades crew logging jobs on site. Twice a year is a bookmark, not an app.
Offline work is the second. Basements, rural roads, warehouses and parkades break web sessions. A technician filling in a form with no signal needs local storage and a sync queue that resolves conflicts later. That is hard to do well on the mobile web.
Device hardware is the third, though the honest version is narrower than the pitch. Mobile browsers reach the camera and location, and an installed web app can receive push notifications. The hardware case holds for background tracking, sustained scanning, and Bluetooth links to equipment.
The fourth is a loyalty loop with money in it. A stored balance, points a customer watches, a repeat order placed in three taps. Loyalty as a marketing word does not qualify. Loyalty as a reason to open the app on a Wednesday does.
What ownership costs after launch
Building is the cheap part. Both major platforms ship a new operating system version most years, and updates can break what used to work. SDKs are deprecated. Signing certificates expire. A quiet app still needs periodic releases simply to stay installable.
Store distribution adds a review process you do not control. An urgent fix can sit waiting for approval. Developer accounts cost money to keep, and platform commission applies to some in-app purchases. Each change means two builds, two test passes and two sets of screenshots.
Then support. Some customers report a bug by posting a public one-star review rather than emailing you. Someone has to read them, reply, and decide what to fix. Plan for maintenance to cost a real share of the build in the first year, and to carry on after that.
An unopened app is worse than none
An icon nobody taps is not neutral. It sits on the home screen you paid to reach, showing last year's prices and last year's design. It ages in public.
Ratings compound the problem. One-star reviews left after a crash two years ago can sit near the top of the listing for anyone searching your name. That is a liability attached to a product you stopped maintaining.
Downloads are the wrong measure. The number worth watching is returning weekly users after the first month. If you cannot picture that number looking healthy, do not start.
Cheaper options worth exhausting first
Fix the mobile site. Load speed on a mid-range phone, tap targets a thumb can hit, a booking form that does not ask for eleven fields. This work serves the customer who found you today, not the one who might install something later.
A progressive web app covers part of the middle ground. It installs to the home screen, caches content for patchy signal, and skips the store. Background capability on iOS remains limited, so test the specific requirement before committing.
Messaging covers much of what businesses actually want an app for. Reminders, confirmations, order updates and follow-ups arrive in a channel people already read. If you sell online, check what your commerce platform's own app and notifications give you before commissioning a build.
A short test before you commit
Write the one-sentence job. Then picture how often one customer does that job in a month. If the honest answer is once or twice, improve the site instead.
Cost year two before year one. Add the build, the store fees, the maintenance releases, and the person who answers the reviews. Compare that total against what the same money would do on the website, on search visibility, or on another pair of hands.
If the case survives that, build the narrowest version that does the one job, on one platform, and watch it for six months. Widening later is cheaper than retreating. For most businesses, the useful outcome of this exercise is a better mobile website.
If a customer would not open it several times a week, the money belongs in the mobile website, not the app store.
Tell us what you want to improve.
A short call to work out what the job is and put a number on it.