Does your small business need an app? A plain decision guide

Short answer: most small businesses do not need their own app on day one. An app pays off when the same customers come back often and want to do one thing quickly on their phone: reorder, rebook, check points or track an order. If your customers visit a few times a year, a fast mobile website does the job for a fraction of the cost. This guide gives you a simple test, the real costs and the order to do things in.

We build and host websites for shops, cafés, clinics and offices in San Jose, and the app question comes up in almost every second meeting. The honest answer is usually "not yet", but there is a point where the website stops being enough. When you reach it, the cheapest sensible route is one codebase for iPhone and Android, which is why many small businesses go with cross-platform Flutter development instead of paying for two separate apps. Below is how to tell which side of that line you are on.

What is the one question that decides it?

Ask yourself: how often does the same customer come back and do the same thing? An app lives on the home screen of people who already like you. It does almost nothing to find new customers, because nobody installs an app for a business they have never heard of. New customers find you through search, maps and word of mouth, and they land on your website. Apps are a retention tool, not a discovery tool.

So the useful number is visit frequency. A coffee shop regular who orders four times a week, a client who books a haircut every five weeks, a family that orders from the same restaurant every Friday: these people will happily install an app if it saves them a minute every time. A homeowner who calls a roofer once every fifteen years will not.

What are the signs your business is ready for an app?

Count how many of these are true for you. One or two means wait. Four or more means it is time to price it out properly.

• Repeat customers make most of your revenue. If 60 percent or more of sales come from people who have bought before, an app gives them a faster way back.
• Ordering or booking is the main action. If most visitors to your site go straight to "order" or "book", an app can cut that from a dozen taps to two.
• You run, or want to run, a loyalty program. Paper stamp cards get lost. A points balance on the phone does not, and it tells you who your best customers are.
• Customers already ask for it. When people at the counter ask "do you have an app?", take it seriously and write down how often it happens.
• You need phone features a website handles poorly. Reliable push notifications, offline use, the camera for scanning, or location for delivery tracking all work better in a native app.
• You pay heavy fees to a third-party platform. If delivery or booking marketplaces take 15 to 30 percent of every order, moving loyal customers to your own channel can pay for development within a year or two.
• Your staff work on their phones. Sometimes the best first app is not for customers at all, but for technicians, drivers or cleaners who need job lists, photos and signatures in the field.

When is a mobile website enough?

For most small businesses, the right first step is a website that loads in under two seconds on a phone, shows hours, prices and location without scrolling, and lets people call, book or buy in a few taps. That covers what most people want from a local business on their phone, and it also helps you show up in local search, which an app cannot do. If you have not done the basics yet, start with our guide to local SEO for small shops.

A website is usually enough when:

• customers buy from you a few times a year or less;
• your main goal is getting found by new customers;
• your booking or ordering tool (for example the one built into your point of sale system) already has a decent customer app;
• you do not have someone who can answer app reviews, update content and plan promotions every month.

There is also a middle step. A progressive web app is a website that can be added to the home screen, opens full screen and can send notifications on most phones. Google keeps a clear overview on web.dev. For a café menu with online ordering or a simple booking flow, a progressive web app gets you most of the app experience without app store fees or reviews.

App, mobile website or web app: how do they compare?

Here is the same decision laid out side by side, using typical small business numbers rather than enterprise budgets.

• Mobile website. Typical build cost: a few thousand dollars, often included in a site redesign. Found in search: yes. Home screen icon: no. Push notifications: no. App store fees: none. Best for: getting found, information, occasional bookings.
• Progressive web app. Typical build cost: roughly 1.5 to 2 times a regular site. Found in search: yes. Home screen icon: yes, if the customer adds it. Push notifications: yes on most modern phones. App store fees: none. Best for: menus, simple ordering and booking, loyalty points.
• Cross-platform app (one codebase for iPhone and Android). Typical build cost for a focused first version: from the mid five figures, depending on features and integrations. Found in search: no, found in app stores. Push notifications: yes, reliable. Store fees: Apple developer program 99 dollars a year, Google Play 25 dollars once. Best for: frequent repeat customers, loyalty, ordering with saved cards.
• Two separate native apps. Typical cost: close to double the cross-platform option, and double the maintenance. Best for: apps that lean hard on device hardware or need the very latest platform features on launch day.

How much does a small business app really cost?

The build is only the first bill. Before you sign anything, put these lines in a spreadsheet for three years, not one:

1. Design and build of the first version. Keep it to one main job (order, book or earn points) plus login. Every extra feature adds weeks.
2. Integrations. Connecting to your point of sale, booking calendar, payment processor or stock system is often the biggest single cost, and the part that breaks most often.
3. Store accounts and reviews. Developer accounts, screenshots, privacy labels and the time it takes for each release to pass review.
4. Hosting and back end. The app needs a server or cloud service for accounts, orders and notifications. Budget a monthly fee, just as you do for website hosting.
5. Updates. Apple and Google release new phone software every year. Plan on 15 to 20 percent of the build cost per year to keep the app working and secure.
6. Marketing the app. Signs at the counter, a line on receipts, a first-order discount. An app nobody installs is the most expensive kind.

The single biggest saving is building once for both platforms. A cross-platform framework such as Flutter lets one team ship the same app to iPhone and Android, so you pay for one design, one set of features and one round of fixes instead of two.

How can you test the idea before you build?

You do not need to spend app money to find out whether customers want an app. Run a 30-day test with what you already have:

1. Set a baseline. Note how many repeat orders or bookings you get per week and how many come through each channel.
2. Make the website feel like an app. Add online ordering or booking, a saved-details checkout and a home screen icon. Our guide to taking payments online covers the payment side.
3. Promote it to regulars only. A QR code at the counter and a short line on receipts: "Order ahead and skip the line".
4. Measure. If a meaningful share of regulars use it every week, you have proof that an app will be used. If almost nobody bothers, you just saved yourself a large invoice.

What should version one of the app include?

A good first app for a small business does one job very well. For a café that might be order ahead, pay with a saved card and collect points. For a salon: see free slots, book, reschedule and get a reminder. For a trades business: request a visit, upload a photo and see when the technician arrives.

Leave out news feeds, chat, games, social sharing and anything you would need to update every day. Add them later only if customers ask. Make sure the first version also covers the boring parts that app stores require: a clear privacy policy, a way to delete an account and an honest description of what data you collect.

How do you choose who builds it?

Ask any developer the same questions you would ask a web host: who owns the code and the store accounts (you should), what happens if you part ways, how updates are billed, and which of your existing systems they have connected to before. Ask to see a live app in the stores, not just screenshots. And ask how they would test the idea cheaply first. A team that pushes you straight into a large build without asking about your repeat customers is selling, not advising.

If you are still at the stage where the website itself needs work, fix that first. A fast, clear site is the foundation every app will lean on anyway, and it is what we do every day at Silicon Data.

Frequently asked questions