Why Microsoft and Google love progressive web apps
- 22 October, 2018 13:00
Progressive web apps just got real.
Though progressive web apps, or PWAs, have been around for about three years — an initiative mostly driven by Google — they got real this week when Google released Chrome 70.
The new version of Google’s web browser comes with a robust roster of new features. But the biggest news is new support for PWAs that work with desktop Windows. (Mac and Linux support should appear in Chrome 72.)
Google and Microsoft compete on many fronts. But when it comes to PWAs, the companies are in perfect alignment. I’ll tell you why below, but first let’s clarify exactly what PWAs are.
PWAs: Easier for all
A PWA is a website that can be made to look and feel like an installed app or application on a smartphone, tablet, laptop or desktop.
PWAs also support or actually replace a mobile-first design strategy, where you can create the PWA for mobile, then make that available on all devices.
Because PWAs bypass app stores, they help solve the problem of app fatigue, where users resist wading into an app store to find yet another app they’ll try once and forget about. When users visit your site, you can offer the PWA installation on the spot, and launch it from that site with every visit.
Most major retailers offer apps that enable loyalty and discount features, as well as a better experience with shopping. But most customers of those retailers have no interest in downloading the apps. PWAs can run when they visit the store, providing additional features that run like regular apps.
Various test cases have proved that PWAs dramatically improve engagement, conversions, interaction, open rates for push notifications and opt-in.
Pinterest launched a PWA designed to replace accessing the service through a regular browser experience. It reported huge benefits, such as a 50 per cent increase in clickthroughs on advertising and a 40 per cent increase in spending by users who spent more than five minutes on the site. The PWA outperformed not only the mobile web usage, but also the mobile app usage.
Also: PWAs support all kinds of devices, including Chromebooks.
The old choice for developers was to create separate apps for Windows, macOS, Linux, iOS and Android — but still fail to serve Chromebooks, unless you created a sixth implementation with a Chrome extension.
The new choice is to create PWAs, and serve all the platforms, including Chromebooks, with a single implementation.
And that same work can ease access to PWAs from smart TVs and other IoT devices. PWAs feel like apps, but the content search engine is indexable and user-shareable.
PWAs are also relatively secure. At installation, they have zero access to the systems’ hardware. This access has to be granted on a case-by-case — a resource-by-resource — basis after explicit permission is granted by the user.
Gaining access to storage, location and Bluetooth requires three separate permissions. Users can say yes to Bluetooth, for example, but turn down the storage and location requests.
This is more or less how mobile apps work, but it’s an improvement on how desktop applications have traditionally functioned. The bottom line is that PWAs finally turn browsers into app platforms — real apps, not yesterday’s horrible web apps.
Why Microsoft is all-in with Google on PWAs
Thanks to Chrome 70, PWAs in Windows 10 function like regular apps. That means they support notifications, Live Tiles and Cortana, and they’re accessible from the Chrome menu, the Start Menu or as a pinned app on the taskbar. And they’re available in the Microsoft Store.
Google and Microsoft are so far totally sympatico when it comes to PWAs. The reason is that PWAs increase the number and range of apps available to Windows users.
But I think the biggest reason is that Microsoft hopes to re-enter the smartphone market with its Andromeda device. Instead of entering a market with no apps, it would instead enter a market with all the PWAs.
Many of these apps will be created primarily as replacements for Android apps. And so many of the apps formerly available only to Android devices and Pixelbook devices will now also be available to Windows Surface Phone devices, or whatever Microsoft ends up calling Andromeda.
It’s a win-win for Microsoft and Google. Microsoft gets tons of apps for its devices. Google gets everybody doing everything from the web, which supports its current ChromeOS strategy and future Fuschia strategy.
Progressive web apps don’t always mean progress
There are downsides.
Discovery for PWAs is decentralised. You can’t just go to an app store and find whatever you’re looking for with a search.
Google maintains a directory of PWAs. But to the best of my knowledge there is no single resource for all the PWAs out there.
What we don’t know is whether the industry can make sure PWAs represent a single app platform, or whether PWAs will either allow or engineer fragmentation. Microsoft and Google have been collaborating on PWAs so far, and that’s a good thing.
Apple, not so much. And while Apple is starting to support PWAs in Safari, it’s not clear that the company is motivated to support common standards. And Safari functionality is wanting. One thing that’s broken for PWAs on iOS is web push notifications, for example.
Raw performance of PWAs is generally lower than native apps. Another downside is that PWAs are highly isolated. So it’s hard and unlikely for different PWAs to share resources or data directly.
So PWAs aren’t perfect. I still think they’re going to be huge.
PWAs are far more efficient for both users and developers. They’re far more flexible, cross-platform and low-footprint than web apps, web sites, mobile apps or desktop apps.
Brands and organisations like Starbucks, Twitter, Burger King, Home Depot and NASA are all switching to PWAs. Maybe your company should, too.
Now that PWAs have arrived for real on Windows, it’s time to get serious about them in your organisation. Conduct an inventory and analysis of all your company’s apps and see which can be converted over to PWAs.
It’s more work, but it will pay off tenfold in the long run.