Every six months or so, a familiar argument flares up on ecommerce Twitter: native is dead, PWAs have won, just ship a fast website. And every six months the data tells a different story. Apps still beat web by a wide margin on session frequency, time on screen, and average order value for repeat buyers. The question for a Shopify brand in 2026 is not whether to build mobile at all — it is whether to do it as a native app, a Progressive Web App, or a hybrid wrapper around your existing storefront.
This piece walks through how each option actually behaves in production, where the differences matter for ecommerce specifically, and the cases where one approach wins decisively over the others. We will skip the architectural philosophy and stay close to the metrics that matter: install-to-purchase rate, repeat purchase rate, push delivery, and how each path holds up under App Store and Play Store policy churn.
What each option actually is
A native app is a binary you submit to the Apple App Store and Google Play. It runs on platform-native APIs, gets its own icon on the home screen, and integrates with system features like push notifications, biometric auth, and Apple Pay or Google Pay without going through the browser. For Shopify, the catalog and checkout are still served by your Shopify backend through the Storefront API, but the UI is built and shipped as a real iOS and Android binary.
A Progressive Web App is a website that adds a service worker, a web manifest, and a few platform hooks to behave more like an app. Users can "Add to Home Screen" on iOS or get prompted to install on Android, and the resulting icon launches a chromeless browser window. There is no App Store listing, no native push on iOS (without workarounds that landed in iOS 16.4 but still suffer delivery quirks), and no platform-store discoverability.
A hybrid wrapper tries to split the difference: a thin native shell wraps a webview that loads your existing storefront. It looks like an app in the App Store but renders web HTML inside. Tapcart, GoodBarber, and Vajro all started this way, and most have since moved toward more native rendering for the screens that matter most. Hybrid is faster to ship than full native, but it inherits the worst parts of both worlds: webview performance, App Store review rules, and a UI that often feels neither here nor there.
Retention is where the real difference shows up
Almost every public benchmark agrees on this: apps see significantly higher repeat-visit rates than mobile web, especially for repeat-purchase categories like beauty, supplements, apparel staples, and food and beverage. The mechanism is not magical. It is the install. An icon on a home screen turns a brand from "something I bought once" into a fixture in the buyer's daily phone routine. They see it. They tap it. They re-buy.
A PWA can achieve a similar effect for users who follow the Add to Home Screen prompt. The catch is that conversion on that prompt is dramatically lower than App Store install conversion. iOS requires multiple taps in Safari, hidden behind a share menu most users do not associate with installing apps. On Android the experience is better, but the install graph still lives outside the Play Store, which means no organic discovery and no review-driven social proof.
Once a user does install, the gap widens. Native push has predictable delivery on both platforms. Average session frequency for installed users on a healthy app sits two to three times higher than the same brand's mobile-web baseline. Cart abandonment recovery via push, in particular, is meaningfully cheaper than the same recovery via email or SMS — partly because there is no per-message cost, and partly because push lands during the exact attention window when the buyer might still close the loop.
When a PWA actually wins
There are real cases where a PWA is the right answer. The most common: very long-tail catalogs where the buyer journey is rare and the brand earns no benefit from being on the home screen. Think large B2B parts catalogs, government-issued forms, one-off-purchase verticals. If your repeat-purchase rate hovers below 10 percent, the install graph does not have much to compound.
PWAs also outperform native when your traffic is dominated by paid social and you have not solved attribution well enough to defend an app channel against tracking erosion. Mobile-web links open in Safari or Chrome, where your existing pixel stack still has some visibility. Native apps land users in a context where Apple's App Tracking Transparency framework limits what you can measure. If your acquisition stack depends heavily on iOS deterministic attribution and you have not yet adopted post-IDFA modeling, do not rush the native build.
Where hybrid leaves money on the table
Hybrid wrappers earned their popularity because they let agencies and platforms ship "an app" in a week. The downside shows up in conversion. A webview-rendered product page often hits 200–400 millisecond slower paint times than a native equivalent on the same device. Multiply that across a category browse plus a product detail plus a cart, and the cumulative friction shows up in funnel drop-off.
The other hybrid tax is App Store policy. Both Apple and Google have tightened rules around webview-wrapper apps that exist mainly to load a website. Reviewers now look for native-feeling UI, native checkout where applicable, and meaningful platform integration. A pure webview shell increasingly draws rejection notes or requests to refactor. The brands that built on hybrid platforms five years ago are quietly migrating off them now, paying the migration tax once instead of fighting the review queue every release.
“The install is not the goal. The install is the moment your brand earns the right to a daily slot on your buyer’s phone.”— Internal Appolar launch playbook
How to choose for your store
The framing we use with brands considering Appolar is simple: native first if your repeat-purchase rate is above 15 percent and you have a 30-day cohort that already returns by mobile web. Below that, a PWA buys you time while you tighten retention through the existing channels. A hybrid wrapper rarely makes sense for a brand that intends to stay on mobile for more than a year; the conversion and review-process tax compounds against you.
Quick test: are you ready for native?
- Repeat purchase rate above 15% on your mobile-web cohort
- A push or SMS list of at least 5,000 engaged subscribers
- Mobile traffic is at least 50% of total sessions
- You have a clear product story to tell beyond your collection grid
- You own (or are willing to set up) Apple Developer and Google Play accounts
If all five are true, native pays back fast. If two or three are true, build the PWA path while you fix the rest. If only one is true, focus on retention basics before adding a new channel — the install graph only matters if the rest of your funnel can absorb it.
Appolar exists to compress the native path: builder in Shopify, push and checkout out of the box, App Store and Play Store submission under your developer accounts. If you are still on mobile web or a hybrid wrapper and the checklist above sounds like you, we can usually have a native build approved by the stores within two to three weeks of kickoff. The harder work is the strategic one: deciding the install is worth chasing, and committing to a mobile channel that compounds.
