For most online stores, a fast mobile-optimised website earns more, because that's where new customers arrive from Google, ads, Instagram and WhatsApp. A shopping app can earn more per customer once you have many loyal, frequent buyers, but it rarely wins new ones. Perfect the mobile store first; add an app when repeat buyers drive real revenue.
The question "which earns more?" hides a trap. Shoppers who install a brand's app are usually its most loyal customers already, so app users naturally look more valuable than website visitors. What matters is the extra revenue the app creates, not the revenue that would have happened on the website anyway. Below is a way to estimate that. If you decide an app is justified, our guide to mobile app development for businesses covers the build itself.
Where does e-commerce revenue actually come from?
Online store revenue is the product of four numbers: visitors, conversion rate, average order value and purchase frequency. A website and an app influence different parts of that equation.
- Visitors: the mobile store wins clearly. It can be found in search and AI answers, opened from any ad or shared link, and needs no install.
- Conversion rate: an app can convert well because users are logged in with saved payment details, but much of this comes from who installs rather than what the app does.
- Average order value: roughly a draw; both can use recommendations, bundles and free-shipping thresholds.
- Purchase frequency: the app's strongest card. A home-screen icon and push notifications can bring loyal buyers back more often.
So the website is your acquisition engine and the app is a loyalty tool. If your growth problem is finding new customers, an app won't fix it.
How do the two compare side by side?
This table summarises the practical differences for a typical small to mid-size online brand. Costs are estimates that vary by scope.
| Factor | Mobile-optimised store | E-commerce app |
|---|---|---|
| Typical build cost | Most website projects US$2k–10k; larger custom stores quoted separately | About US$25k–90k for a cross-platform app, more with complex integrations |
| Time to launch | Up to about 4 weeks for an e-commerce build | 10–20 weeks for a first version |
| Annual upkeep | Hosting, platform fees, updates | Typically 15–25% of build cost, plus store compliance |
| New customer acquisition | Strong: search, ads, social, links | Weak: requires an install first |
| Repeat purchase prompts | Email, WhatsApp, retargeting ads | Push notifications, home-screen presence, plus all of the website channels |
| Checkout speed | Fast with wallet payments and guest checkout | Fast for logged-in users with saved details |
| Store commissions on physical goods | None | None; physical goods use your own payment provider |
| Discoverability in Google | Yes | No; app store search is a separate channel |
Both can accept Apple Pay and Google Pay, so one-tap checkout is no longer an app-only advantage. And because you're selling physical products, Apple's and Google's in-app purchase commission doesn't apply to your sales in either case.
How do you work out whether an app would pay for itself?
Estimate the extra profit an app would create, then compare it with the build and running costs. Here's an illustrative calculation; the figures are hypothetical, so replace them with your own.
| Step | Illustrative figure (USD) |
|---|---|
| App build cost, spread over 3 years | $50,000 ÷ 3 = about $16,700 a year |
| Annual maintenance (about 20% of build) | $10,000 a year |
| Total annual cost of the app | about $26,700 |
| Your gross margin on products | 40% |
| Extra revenue needed to break even | $26,700 ÷ 0.40 = about $66,700 a year |
| Average order value | $50 |
| Extra orders needed per year | about 1,330, or roughly 110 a month |
The key word is extra. These must be orders that wouldn't have happened on the website: customers buying more often because of the app, not existing customers switching channels. If you have 300 loyal customers, finding 110 genuinely additional orders a month is a stretch. If you have 10,000 customers who already buy every month, it may be easy.
A useful rule of thumb: if you can't name the group of repeat customers who would generate those extra orders, you're not ready for an app.
When does a shopping app usually earn its keep?
An app tends to justify itself when several of these are true at once:
- High purchase frequency. Customers buy monthly or more often: groceries, pet supplies, beauty, coffee, fast-moving fashion.
- A large, engaged customer base. Thousands of repeat buyers, not hundreds.
- A loyalty programme worth carrying around. Points, tiers or member pricing that customers check regularly.
- Drops, limited releases or flash sales where a timely push notification genuinely matters to the buyer.
- A big catalogue that people browse for fun, where saved favourites and personalised feeds add value.
- In-store and online overlap, such as scanning products or showing a member card at the till.
Even then, launching the app is only half the job. Getting existing customers to install it takes deliberate effort through email, packaging inserts, checkout prompts and incentives; our playbook on getting your first 1,000 app downloads covers how.
What should you fix in your mobile store before considering an app?
Most stores leave money on the table in their mobile website long before an app would help. These fixes are cheaper and reach every visitor.
- Speed. Compress and convert images to modern formats. On grww.tech we cut portfolio image weight from 1.66 MB to 456 KB by converting to WebP, and load analytics after the page so it doesn't slow the first paint. Product-heavy stores benefit even more from this kind of work.
- Wallet payments and guest checkout. Don't force account creation before purchase.
- Search and filters that work on a phone, including size, colour and price.
- A sticky add-to-cart button and clear delivery costs before checkout.
- Size guides and real photos, which reduce returns as well as hesitation. Fashion brands can find more in our guide to clothing brand websites with lookbooks, size guides and WhatsApp commerce.
- A WhatsApp or chat option for shoppers who want to ask before buying.
- Email and WhatsApp flows for abandoned carts and repeat purchase reminders, which do much of what push notifications would do.
Can messaging and ads do the job of an app?
For many brands, yes, at least until they're much larger. Conversational commerce, where shoppers ask questions and order through WhatsApp or Instagram messages, gives a personal, repeatable channel without an app install.
GRWW's Meta Ads data from a clothing brand shows how far this can go: 2,564 enquiries at an average of ₹5.43 per conversation, with a 2.97% click-through rate. Its best ad drew 640 conversations at ₹3.35 each, and one ad reached a 5.40% click-through rate. Across the accounts we've run, click-to-message campaigns were usually the cheapest way to start a sales conversation. For a brand at that stage, money spent on creative and a fast mobile store is likely to return more than the same money spent on an app.
There's also a middle step: a progressive web app adds home-screen installation and some app-like features to your existing store at a fraction of the cost of a native app.
What about stores that have outgrown their platform?
Sometimes the real issue isn't app vs website but that the store platform itself has become the bottleneck: slow pages, apps-on-apps for basic features, or fees that scale painfully. In that case, rebuilding the store, or moving to a headless setup that can later feed a mobile app from the same backend, may be the better investment. We cover the warning signs in when e-commerce brands should move from Shopify to custom.
Frequently asked questions
Do shopping apps convert better than mobile websites?
App users often show higher conversion rates, but that's largely because the people who install an app are already loyal customers. The fair comparison is how much extra those customers buy because of the app. Measure that incrementally before crediting the app with all their spending.
Do I pay Apple or Google a commission on products sold through my app?
Not for physical goods. App store in-app purchase commissions apply to digital goods and services used in the app, while physical products are paid for through your own payment provider. You'll still pay your normal payment processing fees.
Can I turn my Shopify or WooCommerce store into an app cheaply?
There are tools that wrap a store into an app quickly and cheaply. They can work for testing demand, but they may offer limited customisation, and an app that adds little beyond the website risks rejection or low use. Decide what the app will do better than the site before choosing this route.
How many customers do I need before building an e-commerce app?
There's no fixed number, but you need enough repeat buyers to generate the extra orders your break-even sum requires. For many small brands that means thousands of active repeat customers. Run the calculation with your own margin and order value.
Should I build a progressive web app instead?
A progressive web app is a sensible middle step if you want home-screen installation and some offline or notification features without native app costs. Support for notifications on iPhone is more limited than in a native app. It's often the right choice while your repeat-customer base is still growing.
Deciding for your store
Start with the numbers: your repeat-purchase rate, margin and order value will tell you whether an app adds revenue or just moves it around. If you'd like help running them, or improving your mobile store first, see our recent work and get a fixed-price recommendation.