Why HTML5 Should Replace Native Apps for Ecommerce

Today’s post expands on one of my predictions for 2013 that could be somewhat controversial:

HTML5 will begin to replace native apps in mobile commerce strategies, as Apple’s dominance dwindles and Android and Windows Mobile grow. The ease of maintaining one experience that serves all platforms and the significant cost benefit will make this a no-brainer approach for organizations.

Here’s the rationale behind it. Let’s begin with the cons and pros of native apps:

Cons of native apps

  • Expensive to develop and maintain across multiple platforms (including updates and bug fixes)
  • App users have to proactively update their apps as well, meaning user experience can get worse over time
  • Lack the “linkiness” of the web (e.g. content can’t be shared outside of the native app)
  • Apps that generate revenue are subject to a heft cut by the platform owner (30% for Apple and Google)
  • Consumers don’t use apps as much as you think (will explain below)

Pros of native apps

  • Can leverage properties of a phone like shake and camera, which can add to the cool-factor and utility of your app
  • Having a native app meets customer expectations (e.g. searching for your app in the App Store)
  • Performance can be faster (but some mobile browsers can be faster than native apps due to optimized memory management and performance)
  • New customers / users can discover you in the marketplace through the browse feature

This list is not exhaustive, but you get the idea. There are certainly business cases for building a native app. But my prediction is more businesses will embrace HTML5 apps because of the cost savings and efficiency of maintaining one experience that serves all devices.

Aside from cost, mobile strategists need to take a serious look at the reality of app usage.

Do mobile apps matter for ecommerce?


Mobile shopping is growing, but the question is whether consumers demand native apps or are satisfied with browser experiences. For online retail, research suggests: “across all demographics, 87 per cent prefer shopping via websites and mobile sites, compared to just four per cent that like to use mobile apps.”

According to research by Flurry, mobile shopping through apps isn’t even on the radar. It’s part of the 5% of time spent on the “other” category.

Paid content

Certainly news and magazine content is preferred by tablet owners to be consumed through native apps, right? According to Pew Research, tablet owners are 3 times as likely to access news through mobile web browsers than native apps, and smartphone owners twice as likely.

Flurry’s chart shows consuming news is only 2% of time spent on mobile apps.


If you develop games, tablet and smartphone apps are not accessories to, they are your products. It’s not surprising that games represent 43% of time spent on mobile apps. But HTML5 is still a viable strategy for even the largest game developers. EA’s Strike Fortress was developed in HTML5 by college interns in 5 months.

Social networks

Though a big chunk of the chart, social networking is only 26% of time spent using mobile apps. But even Facebook admits, it gets more visits to its mobile web site than through its Android and iOS apps combined.

Mobile Platform Market Share

If your core customer is affluent, your mobile traffic and conversions may skew towards iPad owners, but market share data from Gartner shows the big player is Android. 72.4% vs. 13.9% is a massive spread.

iOS share is even less than the sum of the rest of mobile platforms, and we may see iOS share drop even further if Android, Blackberry and Windows gain users this year.

But is HTML5 really ready?

Mark Zuckerberg claims Facebook’s “biggest mistake” was doubling down on HTML5. His development team “burnt two years” on their HTML5 experience before shifting gears to native apps, because Facebook believes it can do so much more with native than HTML5.

In a Yammer discussion, my co-workers aptly pointed out this is more likely a case of “a bad workman blaming his tools” than one of HTML5 being inferior. Facebook’s lack of in-house expertise is the real issue.

Sencha Fastbook demonstrates that HTML5 can deliver an even better experience than Facebook’s native apps:

How you like dem apples?

In addition to lack of in-house skills, Facebook seems to have a very backward approach to mobile strategy. Not only is it catering to smaller segment of native app users vs. the web version, it released the new iOS app long before Android. If 13-19% of Facebook’s users are iOS owners, and a much lesser portion prefer to use the native app to the web version, that’s a lot of effort to please the minority.

Bottom line

User behavior is geared more to web apps than native, and the business case for developing dedicated apps for multiple platforms is losing steam. Add to this that HTML5 has been deemed “feature complete” by the World Wide Web Consortium. My prediction is mobile strategies that consider the ROI of developing apps, with a long-term outlook will shift away from native development, though it may take some time to catch on, as the hype around native apps has convinced many marketing departments native ecommerce apps are table stakes.

Tags: , ,

Related Articles

14 Responses to “Why HTML5 Should Replace Native Apps for Ecommerce”

  1. Jesse says:

    Your list of pros and cons is very sensible, and your the article’s reasoning is very sound as long as it sticks to those. However, I think you may be oversimplifying the marketshare issue with iOS, as well as making big assumptions about Facebook and its users.

    I think there is probably an issue of scale that factors into the “native vs. HTML5″ discussion. For a great majority of ecommerce sites, a native app for any platform is probably redundant. My (admittedly anecdotal) experience is that people are very unlikely to download an app for a site they are using for the first time. Given that, the site’s web experience had better be good enough to satisfy their needs or they won’t be back! If the site meets their needs, why spend the resources developing a native app?

    HOWEVER. If you are a really well-known brand, and a site that people use daily (or more!), you have to consider a few additional things.

    #1: A user of a mobile site won’t mind a delay of a second or two nearly as much if they only visit the site occasionally (or once). If they use your site frequently, it will start to be worth their while to download an app if it shaves off some of those precious load times. For example, Amazon is my first stop in any online shopping, even if I don’t end up buying there. When I’m on my phone, I can pop open the app and do search after search without experiencing lag. I don’t find that to be the case with their website, as good as it is!

    #2: iOS users love their devices, and they do more online shopping than Android users on average, especially on the iPad (http://www.asymco.com/2012/11/26/the-android-engagement-paradox/). There is a potential advantage in getting them to view your site alone in an app rather than having it open in a tab along with competing shopping sites.

    #3: Going along with #2, there seems to be a perception among mobile users that “serious” sites will have native apps. Whether it is fair or not, Facebook may feel that they need to maintain a native app lest they look like they aren’t serious about the mobile experience. And just look at the Android community’s reaction to any major company’s iOS app. They aren’t saying “why did they make a native app?” They are saying “we want a native app, too!”

    Again, this mostly applies to well known sites with strong existing engagement from customers and users. I personally am sick of every site with a forum prompting me to download their stupid app (no, I don’t want to download an app so I can visit the “video card geeks” forum over and over — I just want to fix my video card!) I am all in favor of wider adoption of HTML5 for mobile, but you may be overstating your case when you try to judge Facebook’s developers without being privy to all of their technical, social, and marketing concerns.

  2. John Back says:

    The “much more” that Facebook can do with native may have even boiled down to just the links, which they effectively portalised for any user not willing to go through the two-step ‘hit link + view outside app’ process. I’m sure the additional ‘big demographic data’ they generate through this is already paying dividends for their ad teams.

  3. I have to agree with your view of avoiding “native apps” as a means of simplifying things for the consumer end user. I would like to avoid what happened in the very early days of computer time-sharing, where all an online user needed was a “Dumb” terminal to connect with an online application. Then, the PC came out and killed the concept of sharing anything.

    With HTML5, along with WebRTC, we have all the ingredients for mobile business, including both multi-modal self-service applications, CEBP, and unified communications .

  4. Alex Moseman says:

    These types of articles have been going around alot lately. It always seems to me that they make it a black and white conversation. The truth is that the different products support different user types and use cases. No one is going to go from an email or a search link to an app. By comparison, a heavy loyal user looking for increased functionality specific to a context (in-store, at home, walking by a store) is looking for a very different and native experience. To put it simply, Mobile web will support 80% of a brands customers, where an App is better suited for the 20% of the loyal brand users.

    If all you want to do is allow a user to find a product or products, over a reliable internet connection, and complete a purchase, then a browser experience should be fine. But if that is all you want to do, then you are probably missing some important opportunities for both your brand and your customers. In general, you should never let the technology define the requirements.

    • Hi Alex, I understand what you’re saying. The article really intended to predict a trend away from developing native apps in favor of HTML5 rather than a mobile website vs. mobile app debate. If you have a business case for an app, the decision needs to be made about what platforms to develop for, whether you want to develop device-specific, web app or both. There’s no black and white right or wrong approach, but my prediction is there will be less native app focus going forward, because of the “readiness” of HTML5 and the resource savings build-once-serve-all offers, whether this shift happens this year or in the next few years.

  5. Kevin says:

    1) Not for ecom items. Only if you charge for the app: “Apps that generate revenue are subject to a heft cut by the platform owner (30% for Apple and Google)”

    2) Why do you think so? Can hook up social sharing tools to Native apps. “Lack the “linkiness” of the web (e.g. content can’t be shared outside of the native app)”

    • Hi Kevin, yes I was referring to paid apps such as subscription/in-app virtual goods, not transaction value.

      Re: linkiness, again this refers to paid content industry (news, magazine, ebook) rather than product page content. This is something that’s come up frequently at conferences and in articles on why consumers are unsatisfied with paid content apps. Digging a bit deeper it seems the problem is not that apps can’t link out, it’s that existing content apps are not doing them well or it’s not intuitive for users. I’ll edit that point from the list of cons.

  6. Philip says:

    Android may dominate in terms of OS market share, but if you look at the mobile browsing and mobile shopping breakdown, iOS dominates.

    For instance, elf cosmetics reports that iPhones make up 70% of all mobile traffic and 79% of mobile sales, compared to all Android devices combined that contribute 21% of mobile sales. (http://www.internetretailer.com/commentary/2012/12/17/apple-users-are-far-more-valuable-android-users)

  7. Damian D. says:

    HTML5 was considered a big waste of time by Facebook, and its smartphone apps reflected it. It took a lot of time from their developers and the resulting product was quite subpar. Still, HTML5 does provide some interesting applications for certain projects. It is important to evaluate first whether a project should be an app or a mobile site. Thanks for providing the list of pros and cons because they will provide a useful checklist for web developers.

  8. As a long time fan of open source software, I’m pleased to see Android dominating so heavily. I’m also glad that Apple’s strategy of locking everything down and trying to control everything on a user’s phone doesn’t seem to have paid off.

    Also very surprised to see that most people use facebook via the mobile version rather than the app. I’m pretty sure that a quick poll of my friends would find nearly 100% using the app instead.

  9. For Mobile commerce native apps should be the default answer for one little, and yet very important reason. Native apps tend not to depend on network connection that much and therefore cannot be slowed down as much as a web site can. If market research shows that even 1 additional second of friction in the checkout process can cost a vendor 7% in conversions (completed transactions) – then said vendor should really put speed as a priority and therefore put all HTML5 etc. in the proverbial drawer.

  10. Suana William says:

    From development point of view , Web apps have an advantage over native apps. Users dont prefer an app that has updates every week.
    For e-commerce where contents and things needs to be changed frequently I prefer to use web apps over native apps. for more detail visist at http://www.andyskipper.com/category/ecommerce/

Leave a Reply

© 2014 Get Elastic Ecommerce Blog. All rights reserved. Site Admin · Entries RSS · Comments RSS