PWA vs Native App: What a Small Business Should Build
Written by Kamran, Founder, 3xRevenue. Published · 10 min read.
Most small businesses should build a progressive web app (PWA) first. It works on any phone from one link, skips Apple's $99 yearly fee, and cost about $14,000 less over two years in our example. Build a native app only if you need Bluetooth, NFC, background tasks or an App Store listing.
Here is the moment that starts this question. A customer asks, "Do you have an app?" and you wonder if you are losing bookings without one. Then a developer quotes two separate apps, and the number makes you wince.
You do not need to guess. Ask what your customers do on their phones, then match the build to that job. Need a working first version in 4 weeks? Our fixed-fee MVP sprint ships it as a web or mobile app.
Which one should a small business build first?
Build a PWA first if your customers book, order, pay, check status or read updates. Those jobs need a screen, a form and a database. A browser handles all of them well in 2026.
A PWA is still a website. It lives at your web address, so Google can find it and a text message can link to it. Customers can also save it to their home screen, where it opens full screen like an app.
A native app is built for iPhone and Android and sold through the App Store and Google Play. Today most business apps use one cross-platform codebase for both stores. That is cheaper than two separate apps, but it still means store accounts, store reviews and store rules.
The honest rule: pick the PWA unless a need in the next section is real today. "Maybe later" is not a reason to pay for a store app now.
What can a PWA do on an iPhone in 2026?
A PWA can do most of what a service business needs on iPhone, with a few hard gaps. The biggest gap is install. iPhone users must tap Share and then "Add to Home Screen". There is no install pop-up like on Android.
Push alerts are the second limit. Since iOS and iPadOS 16.4, Home Screen web apps can ask for push permission. WebKit says only a web app "that has been added to the Home Screen" can ask. The request must follow a tap, so you cannot ask on page load.
| Feature | PWA on iPhone | PWA on Android | Native or cross-platform app |
|---|---|---|---|
| Open from a link, no download | Yes | Yes | No |
| Install to home screen | Manual, via Share | Install prompt | Store download |
| Push notifications | Yes, after Home Screen install | Yes | Yes |
| Badge count on icon | Yes, after Home Screen install | Yes | Yes |
| Camera and location | Yes | Yes | Yes |
| Bluetooth and NFC | No | Yes, in Chrome | Yes |
| Background sync while closed | No | Partly | Yes |
| App Store or Google Play listing | No | Possible with a wrapper | Yes |
| Ship an update | Instantly, next page load | Instantly, next page load | After store review |
| Yearly store fees | $0 | $0 | $99 Apple, plus $25 once for Google |
Source: WebKit blog, Apple Developer Program and Google Play Console Help, checked October 2026.
Check your site speed and setup free
A slow website makes a slow PWA. Run the free check above before you build. If your current site loads in 6 seconds, fix that first, because the PWA sits on top of it.
When is a store app worth the extra cost?
A store app is worth it when your product depends on the phone itself, not just a screen. These are the cases where we would skip the PWA:
- You talk to hardware. Door locks, scanners, scales, fitness bands and printers often need Bluetooth or NFC.
- Work happens while the app is closed. All-day driver tracking or large background uploads need native access.
- Push is the whole product. If daily alerts drive revenue, you cannot wait for iPhone users to find Share.
- Customers search the stores. Your buyers look for services like yours in the App Store, not on Google.
- You sell digital content. Paid courses or subscriptions inside the app follow store payment rules and store fees.
If none of these fit, a store app mostly adds cost and review time. Apple also pushes back on thin apps. Its guideline 4.2 asks for features that "elevate it beyond a repackaged website." A store app that only wraps your site risks rejection.
When a store app is the right call, one codebase for both stores is the usual path. Our 6-week mobile app sprint builds iPhone and Android from one codebase for $30,000 and helps with store submission.
Do app store fees apply to bookings and payments?
No, not for real-world services. Apple's App Review Guideline 3.1.3(e) covers services "consumed outside of the app." For those, apps "must use purchase methods other than in-app purchase," such as Apple Pay or card entry.
That means a salon, clinic, cleaner or contractor pays normal card processing on bookings. You do not pay Apple's 15% or 30% cut on a haircut or a deep clean.
The store cut does hit digital goods. Apple's Small Business Program charges 15% for developers with up to $1 million in yearly proceeds. Google Play charges 15% on the first $1 million in most markets, per its service fee rules. So a fitness studio selling video classes in its app pays the cut. The same studio taking class bookings does not.
| What the customer buys | Example | Store fee on the sale |
|---|---|---|
| A service used in the real world | Haircut, cleaning, repair visit | None, use card or Apple Pay |
| A physical product | Coffee beans, parts, clothing | None, use card or Apple Pay |
| Digital content inside the app | Video classes, premium features | 15% for small developers, up to 30% |
| A subscription to app content | Monthly members area | 15% on most small-developer plans |
Source: Apple App Review Guidelines, Apple Small Business Program and Google Play Console Help, checked October 2026.
The math for a 2-van cleaning company
This is an example, not a real client. A cleaning company with 2 vans wants online booking and deposits. Customers should also see arrival windows and rebook in one tap. Both builds use our published fixed prices.
The PWA version is the 4-week MVP sprint at $18,000. The store version is the 6-week mobile app sprint at $30,000. Add Apple's $99 yearly fee and Google's $25 one-time fee. For upkeep, we use 15% to 20% of the build cost per year. That is a common industry estimate, quoted by app firms such as Cleveroad, not a fixed price.
| Cost item | PWA | iPhone and Android app |
|---|---|---|
| Build | $18,000 | $30,000 |
| Store fees, year 1 | $0 | $124 ($99 Apple + $25 Google) |
| Upkeep, year 2 (15% to 20%, estimate) | $2,700 to $3,600 | $4,500 to $6,000 |
| Store fees, year 2 | $0 | $99 |
| Two-year total | $20,700 to $21,600 | $34,723 to $36,223 |
Source: 3xRevenue service pricing, Apple Developer Program enrollment page, Google Play Console Help and Cleveroad app maintenance guide (2026). Upkeep figures are estimates.
The gap is about $14,000 over two years. For this company, nothing on the store-app list applies. Bookings are a real-world service, so there is no store cut either way. The PWA is the clear pick.
Now change one fact. Say the company wants cleaners to scan an NFC tag at each job to log arrival. That needs NFC, which iPhone browsers do not support. Now the store app earns its extra cost.
Six questions that settle PWA vs native
Answer these in order. One clear "yes" in questions 1 to 3 points to a store app. Otherwise, start with a PWA.
1. Does the app need Bluetooth, NFC or other hardware?
List every device the app must talk to. Example: a gym's door scanner needs NFC, but a booking calendar needs nothing. Quick check: if the list is empty, cross this off.
2. Must it work while the app is closed?
Write down anything that should run in the background. Example: live driver tracking needs it, while a "your cleaner is 20 minutes away" text does not. Quick check: could a text or email do the job instead?
3. Do customers find businesses like yours in the app stores?
Search the App Store for your service and your city. Example: if only big national platforms show up, local firms are not winning there. Quick check: no local rivals in the store means a weak case for a listing.
4. How often do customers come back?
Count visits per customer per month. Example: weekly gym members may install an app, but a once-a-year roofer customer will not. Quick check: under 2 visits a month favors a link they can open anytime.
5. How often will you change it?
Plan your first 6 months of updates. Example: a new price list each season ships instantly on a PWA. A store app waits for review, which Apple says takes under 48 hours for 90% of submissions. Quick check: weekly changes favor the PWA.
6. What can you spend in year one?
Add build, store fees and upkeep, as in the table above. Example: an $18,000 budget fits a PWA but not a two-store app. Quick check: keep 15% to 20% aside for the second year.
Mistakes that waste an app budget
- Building two native apps first. One cross-platform codebase covers both stores. Two separate apps mean two codebases to build and fix.
- Wrapping your website and calling it an app. Apple may reject it under guideline 4.2, and customers notice.
- Forgetting the home screen icon. A PWA needs icons in several sizes. Our free favicon and app icon generator makes them in a minute.
- Asking for push on the first page load. iPhone blocks the request unless it follows a tap. Ask after a booking, when the alert has a clear reason.
- Rebuilding scheduling from scratch. If a booking link is enough, read our Calendly vs Cal.com breakdown first. A custom calendar only pays off when the tools fall short.
- Skipping the account handover. Store accounts, domains and code should be in your name from day one.
FAQ
Can a PWA be listed in the App Store?
Not directly. You can wrap a PWA in a native shell and submit it. But Apple wants apps that go beyond a repackaged website. Plan real app features if the store listing matters to you.
Do PWAs work offline?
Yes, for the screens and data you choose to cache. A customer can open a saved booking or a price list without signal. New bookings still need a connection to reach your system.
Is a PWA cheaper to maintain than a native app?
Usually yes. You maintain one web codebase, pay no store fees and skip store review for each update. In our example, year-two upkeep was $2,700 to $3,600 for the PWA. The store app needed $4,500 to $6,000.
Can I start with a PWA and switch to a store app later?
Yes, and that is the path we recommend for most service businesses. The database, login and business rules carry over. You add a store app only when one of the six questions turns into a clear yes.
Do Google's new Android rules affect a PWA?
No. Google's developer verification covers Android app packages, with a global rollout planned for 2027. A PWA runs from your website, so it is not an app package.
Start with the version your customers will open this week
Most service businesses do not need the App Store to win bookings. They need a fast, clear phone experience that customers can open from a text or a Google search. A PWA gives you that for less money and ships updates the same day. If your product truly needs hardware or background work, build the store app from one codebase. Either way, a fixed scope keeps the cost from drifting. Tell us what customers must do on their phones. We will scope it in our 4-week MVP sprint.