HTML5 or native mobile app? How about both?

1Executive Summary

When the Financial Times suddenly announced earlier this year that it was switching mobile gears from a native iOS app to build a web app, many in the industry took it as a sign that the dominance of the native mobile app was coming to an end. The paper’s choice had much to do with its position as a subscription-based business. While it shows that high-profile apps are going around important distribution channels like Apple’s App Store, the move doesn’t necessarily spell the the decline of the native app.

The benefits of HTML5

There’s no denying that HTML5-based mobile apps are on the rise. Web giants like Amazon, Google and Microsoft have put their weight behind the emerging standard, and the number of companies seeking to hire developers who are familiar with it is on a steep upward trajectory. Web development platform maker Sencha polled its 2-million-plus developers in June and found that of those interested in building mobile apps, two-thirds were interested in HTML5 and just 6 percent in native mobile technologies like iOS, Android, Windows Phone 7 and BlackBerry OS.

And it’s no secret why: There are plenty of benefits to building web-based mobile apps:

  • It reduces fragmentation for developers. Developers that build native apps and want to cover several mobile platforms were, up until recently, forced to have separate teams working on the iPhone/iPad, Android phones, Blackberry devices and Windows Phone 7. With HTML5, the web is the same on every mobile device. That means you can build one app and know it will immediately work on a variety of devices from different manufacturers.
  • It can reduce cost and time of development, and it’s easier to hire. When you reduce fragmentation, it ends up saving money and man-hours. You can have one team working on the web app rather than devoting resources to an iOS team and Android team and so on. And when it comes to potential hires who can build apps for you, the pool is much larger. Points out Aditya Bansod, the senior director of project management for Sencha, “The number of people who understand how to build web content is between 5 [million] and 10 million, compared to iPhone and Android developers, which is not over 100,000.”
  • Once on the phone, apps function fairly similarly. Some people may not notice (nor care) that their favorite news reader app or the gateway to their business travel tracker app isn’t native. Native apps appear as a button on-screen that you tap to launch. The same goes for a web-based app, which can be saved as a shortcut on the home screen and launched just like a native app.
  • Building web-based apps allows developers to keep more revenue. When developers distribute apps via the web, they cut out Apple or other app store proprietors from getting a part of their sales. Apple, for example, takes 30 percent of all purchases made on or through the App Store.

The type of app matters

Despite all the positives, choosing to make your mobile app solely in HTML5 is not necessarily for everyone. The type of app you’re building is crucial in the evaluation of whether to choose native or HTML5. Richard Rabins, the co-chairman of AlphaSoftware, a development platform for building HTML5 web apps, argues that the biggest opportunity for HTML5 in mobile is for apps not well-represented in the top two app stores, Android Market and the iOS App Store. Specifically, he points to business apps for large corporations. “Management apps, industry apps that connect companies with suppliers buying insurance, filing claims, etc. There are hundreds of thousands of apps in [app stores] today, and relatively few are business apps,” he says. Another example of an underrepresented native app category is software for universities. Apps for registration, tuition payment, class scheduling and email access all need to be easily available on the web but also on as many devices as possible, since a student population is guaranteed to have a wide variety of mobile devices. And when you can build an app once in the same programming language, such as HTML5, it saves time and money.

It’s true that there have been some high-profile switches from Apple’s App Store to apps built on HTML5 and distributed through the web recently, including the aforementioned Financial Times’, and some important debuts, like Amazon’s Cloud Reader, among others. But those changes can be attributed to the specific needs of those developers.

The Financial Times, for instance, wanted to keep subscriber information once people signed up for the app — something that’s standard in the publishing business — from its app, which Apple does not allow publishers to keep for customer privacy reasons. It’s certainly a bonus that as a result of the move the FT won’t have to share any of its revenue with Apple. But more importantly, the Financial Times can get away with a move like this without many adverse effects to its bottom line because it doesn’t really need Apple. Its brand stands on its own, and the publication can pick up mobile readers without being searchable in the App Store directory.

The story is similar with Amazon, which has a business that competes with Apple’s iBookstore. Amazon too is smart to build its own way around Apple’s store yet still have a way to be used on the extremely popular iPhone and iPad, just through the browser.

Rabins says, “It was the equivalent of an atomic bomb when the FT announced they switched from native to HTML5 because they felt like they could give the same experience and wouldn’t have to share subscription revenue.” Still, he said he’s “not sure” if the rise of HTML5 will make much of a dent in the Android, Apple or other native app stores necessarily, because, as mentioned earlier, the reason the Financial Times and some other big names switched isn’t something all developers are going to want to necessarily follow.

What HTML5 can’t touch

However, mobile web apps are limited in ways that native apps are not. Speed is the biggest issue right now, admits Sencha’s Banson. “The browser is an intermediate layer for hardware on mobile and acts as a performance barrier,” he said. “The biggest gap people see tends to be performance characteristics, like how fast can I flick a list.” It can be subtle, but he said it’s “something that people notice.” Over time he expects this to improve.

Accessing some phone features can also be a problem, at least right now for basic mobile browsers. For instance, if you’re making a web app for the iPhone, you can’t access the camera or the in-app payment system. So that means no photos or media-based apps need apply, nor augmented-reality apps, which also use the camera. And if you’re a small development shop without access to Apple’s integrated payment system, it can be more complicated for customers to pay for items than just clicking a button within an app and charging their iTunes account. Granted, these shortcomings won’t necessarily be with us forever. Mozilla, for example, is working on a camera API for its browser. And just because features are not available currently doesn’t mean they won’t ever be: Case in point, access to the iPhone’s geolocation tools and accelerometer were not available in iOS 2.0 but were added by version 3.0.

A happy medium

Others say there are user interface details that, when launching a web app on a mobile device, are just plain annoying. As Stewart Putney, the CEO of social games maker Moblyng, puts it, “just two extra clicks can be like the death of businesses.” His solution: either build a native app to begin with or “wrap” HTML5 apps in native code so that they can be purchased through an App Store.

There are development platforms popping up that help app creators do this and that can render the discussion of web versus native — and what seems on the surface like a black-and-white choice — unnecessary. The kinks are still being worked out while it’s in beta, but Strobe is a perfect example of a nice, middle-ground platform. It allows developers to build an app in HTML5 and other web technologies and then distribute it as both a web app and a native app. It solves the problem of choosing a platform and can still save time and resources on the development end.

Combining the two platforms is an area that’s still in early days, but it’s a good sign of where things are heading. Or as Strobe founder Charles Jolley put it in an interview with Om last year, “People are looking for an either/or solution, but it is not going to end up like that.”

This is true. Smartphone usage is only going to grow as prices go down, the technology matures and mobile broadband penetration rates go higher. And the demand for mobile apps will grow right along with it. In order to take advantage of that and make their apps available to the most users at once, as efficiently as possible, developers are wise to consider solutions that companies like Strobe are building.

Relevant analyst in applications
You must be logged in to post a comment.
15 Comments Subscribers to comment
  1. Thanks so much for writing this. I’ve been having quite a number of discussions with people about how a lot of the reporting on this lately has been an “either/or.” The FT foray into HTML 5 vs. native app really brought a critical question to light: do you NEED to make a native app? If you are a news-reading site, the browser’s ability to save locally should be all you need. You don’t need access to the camera or the GPS or other device elements. And, as you pointed out, companies like Strobe (and Apptanium) are providing frameworks to do both. Great stuff.

Explore Related Topics

Learn about our services or Contact us: Email / 800-292-3024