· 6 min read
Web app or mobile app first: how founders decide
Web app or mobile app first? How to decide from where your users are, what the phone must do, how people find you, and when a PWA is enough to start.
Updated

Whether to build a web app or mobile app first is one of the earliest decisions a founder makes, and it shapes the budget, the timeline and how quickly you can learn from real users. Many founders assume a startup needs an app in the stores from day one. Often it does not. Sometimes it does, and starting on the web would be a mistake.
In short: build a web app first if your first users are at a desk, arrive from links and search, and the core feature does not need the device. Build a mobile app first if the product depends on background location, Bluetooth, device files, content from other apps, or notifications people act on within minutes. If the two sides of the product live on different devices, build both on one API.
This guide sets out the questions we use to decide, what a progressive web app can and cannot do, and the signs that a phone app is essential from the start.
Start with where your users are
Picture your first user at the moment they need your product. Where are they, and what are they holding?
- At a desk, doing work. Dashboards, admin tools, reporting, anything with long forms or tables. This is web territory, and many business buyers will judge your product on a laptop screen. Serpzeno, our own SEO audit tool, is this kind of product.
- On the move, for a minute at a time. Booking, checking in, logging something as it happens, messaging. This leans towards the phone.
- Both, but in different roles. Marketplaces often split this way. Landlords or sellers manage listings at a desk; tenants or buyers browse on their phones.
Most products have one moment that matters more than the rest. Build the first version for that moment.
Ask what the phone has to do
A modern web app can do far more than most founders expect. It can use the camera, ask for the user's location, take payments with Apple Pay and Google Pay, and work on a phone screen as smoothly as on a laptop. 6lyets, an online store we built, takes cards, Google Pay and Apple Pay in the browser.
What the web cannot do as well is anything that needs the phone to work in the background, or to talk closely to the device itself. If your core feature depends on one of these, you probably need a phone app:
- Background location, such as tracking a delivery or a run while the screen is off.
- Bluetooth hardware, such as a sensor, a lock or a printer. Safari on iPhone does not support Web Bluetooth.
- Deep access to files on the device. Our own Headroom storage cleaner scans the phone's files and compares duplicates byte by byte. That could only ever be a native Android app.
- Receiving content from other apps. Lead Diary captures customer enquiries shared from chat apps, screenshots and voice notes, and works offline with no account. Being on the phone is the product.
- Reliable notifications as the main feature, such as reminders or alerts that people act on within minutes.
If none of these is central to your first version, a web app is likely the faster and cheaper place to start.
What can a progressive web app (PWA) do?
A progressive web app, or PWA, is a website built to behave like an installed app. MDN describes it as an app built with web technologies that provides a user experience like that of a platform-specific app, and its guide to progressive web apps is a good introduction. In practice, a PWA can:
- Be added to the home screen with its own icon, and open full screen without the browser bar.
- Keep working offline, or on a poor connection, for the screens you choose.
- Send push notifications. On iPhone and iPad this arrived with iOS 16.4, and only for web apps the user has added to their home screen, as WebKit's announcement explains.
- Update instantly, with no store review, whenever you deploy.
The limits are real too. A PWA is not listed in the App Store, installing one on an iPhone takes a few taps through the Share menu that many users have never seen, and background work is far more restricted than in a native app. For an MVP, those limits often do not matter. For a consumer product that lives or dies by daily notifications, they might.
Think about how people will find you
If most of your users will arrive from a search result, a shared link, an email or an advert, the web has a clear advantage: they click and they are in. Every extra step, such as going to a store, downloading and creating an account, loses a share of people. A content product such as Exode20, a blogging platform we built where readers browse posts by tag and join the discussion, belongs on the web for that reason.
If your users already search the app stores for products like yours, or your market expects an app icon on the home screen, a store listing is part of your marketing, and a web-only launch may look less credible.
Count the cost and the speed of change
A first version is about learning quickly. The web helps in two ways:
- One codebase. A single responsive web app serves phones, tablets and laptops. A mobile app built with React Native or Flutter covers iOS and Android from one codebase, but usually still needs a web admin panel alongside it.
- No review queue. A fix on the web goes live the moment it is deployed. A store app goes through review, and users update in their own time. Our app store launch checklist shows what a store release involves.
Neither of these is a reason to avoid mobile if the product needs it. They are a reason not to build an app just because it feels more serious.
Web app or mobile app first: three common answers
Web first, designed for phones. This is the most common answer for B2B tools, marketplaces in their early days and products that grow through links. Design every screen for a phone from the start, so moving to an app later is a change of wrapper, not a redesign.
Mobile first. When the product only makes sense in a pocket: field work, fitness, scanning, anything built on the camera, location or hardware. Choose a cross-platform framework unless one platform and deep device access settle it for you. Our comparison of React Native vs Flutter covers that choice.
Both, sharing one backend. Sometimes the two sides of a product really do live on different devices. The key is a single API that serves both, so the second client is cheap to add. Rehnaa began as an Expo mobile app. We then built its web version on the same API the app already used, with every app screen for tenants and landlords, and a layout that switches to the app's bottom tab bar below 1024 pixels.
A short decision checklist
- Where is the first user when they need the product: at a desk or on the move?
- Does the core feature need background location, Bluetooth, device files or other apps?
- Are notifications the product, or a nice extra?
- Will users arrive from links and search, or from the app stores?
- How often do you expect to change the product in the first three months?
- Who else needs a screen: an admin, a seller, a support team?
If most answers point to a desk, links and frequent change, build for the web first. If the core feature needs the device, build the app. If you are torn, build a web app that works beautifully on phones and let your first users tell you whether they want it on their home screen.
We make this call in the blueprint week of every project, before design starts, and write the reason into the plan. You can read how we approach each route on our web app development and mobile app development pages, or tell us about your product and we will recommend where to start.
- Product strategy
- Mobile apps
- Web apps
Building a first version of your product?
We scope it in a week and launch it in 8–12, for a fixed price. Tell us what you want to build.
Keep reading

· 8 min read
How to build a doctor appointment app: lessons from DocRadar
Doctor appointment app development, from a Flutter app we built: real-time availability, doctors nearby on a map, in-app payments, roles and store review.
Read the guide
· 7 min read
PoC vs prototype vs MVP: which should you build first?
PoC vs prototype vs MVP: what each one proves, who it is for, what it costs in time, and a simple way to decide which one your startup should build first.
Read the guide
· 8 min read
How to build a marketplace app: lessons from Rehnaa
How to build a marketplace app, from a rental marketplace we built: two kinds of user, trust, map search, chat, one API for app, web and admin, and testing.
Read the guide