The Mobile Dilemma
App or Website? A look at stats, strategies and scenarios
Check your email. Find a restaurant. Compare a price. Schedule a meeting. Research a competitor. Watch a movie, listen to music, read a book, follow a stock, post to a blog, find a recipe, etc., etc., etc. For just about everything you do in life or work, there's an app or a website you can hold in the palm of your hand.
Mobile, particularly as manifest in smartphones, is quickly moving to the center of the online ecosystem. According to Nielsen, the majority of U.S. mobile subscribers will be carrying a smartphone as soon as the end of 2011. 1The Nielsen Company, “The State of Mobile Apps,” www.nielsen.com Gartner projects smartphone sales to more than triple by 2012, to 563.8 million from 172.4 million in 2009. And once users get a taste of always-anywhere connectivity, they're quickly hooked and would sooner run back home for a forgotten phone than a forgotten wallet.
The Dawn of the App Age
Users have been able to access the Internet and certain types of applications from their feature phones for many years. But the iPhone and the Apple App Store unleashed an explosion of mobile computing power and connectivity, opening the door to unimagined functionality and convenience. "There's an app for that" became the mantra of the connected cadre.
But the rise of smartphones and apps — now no longer the exclusive domain of Apple — also introduced the age of mobile Web browsing. With every smartphone platform now supporting a bona fide Web browser, and touch screens maximizing viewable area, virtually the entire Internet is now the purview of handheld devices. But unlike the desktop, mobile users have two major avenues to access online content: apps and the Web browser.
The Marketer's Dilemma: App, Website, Both?
A mobile strategy is no longer optional or an item for the long-range plan. It is a vital, critical channel right now, and increasingly more so every day; marketers ignore it at their peril. If any doubts remain, consider eBay, which expects in 2010 to do $1.5 billion in global sales through phones and tablets — double its 2009 mobile total — and Amazon, which is looking to do more than $1 billion in 2010. By 2015 — four short years away — the mobile commerce market is expected to exceed $119 billion. 2Fast Company, “EBay Dials M for Makeover,” by Dan Macsai, December 2010/January 2011
It's not so simple as the nascent days of the wireline Web, though, when the choice was only to do a website and how fancy to make it. In the early iPhone days, apps were the marketer's golden ring, shiny and prestigious, a badge of forward-thinking techno-savviness, whether or not they moved the bottom line. Today, however, apps aren't quite the status symbol they used to be; but they've also grown beyond their novelty to become useful tools for users and valuable brand and sales channels for marketers. At the same time, the mobile Web has become at least as important, or more so, in the mobile ecosystem.
So the question of the moment is app, website, or both? Apps have a clear edge in caché and potential "coolness" factor, but require multiple builds and maintenance for each platform. Websites are enticing, with their potential to build once for all platforms, but with an overabundance of devices and software, that promise is still problematic. A hybrid approach combining app(s) and website(s) is the gold standard, but expensive and labor-intensive.
How do you arrive at a viable mobile strategy? Benchmark has rounded up some useful data to help you decide.
What Consumers Prefer: Surveys Say
Unfortunately, if you are looking to consumers to help solve the app-or-website dilemma, they're not going to be much help. Some demographic trends are fairly clear, but overall, consumer preference is not overwhelmingly in one camp or the other. Use cases drive many preferences. The same goes for developers and marketers.
A 2010 study by Keynote for Adobe Systems of 1,200 U.S. consumers found roughly equal satisfaction with app and browser experiences, and that users spend about an equal amount of time with each. Two-thirds expressed a preference for browsers when accessing media and entertainment and consumer products/shopping, though again, satisfaction in these categories was neck-and-neck for both websites and apps. Categories where users in the study showed a clear preference for apps were games, social media, maps and music. 3Adobe Systems, “Adobe Mobile Experience Survey: What Users Want from Media, Finance, Travel & Shopping,” October 2010
Inclusive of both apps and websites, these are the top mobile activities as a percentage of users surveyed:
|Accessing social media:||76%|
|Finding local info (events, weather, reviews, maps, etc.):||73%|
|Reviewing bank account info:||67%|
|Listening to music:||63%|
|Playing games more than once/day:||61%|
In Europe, preferences are clear, at least according to a study by Orange. Orange Mobile Exposure 2010 reports that 70 percent of mobile media users in the UK prefer the browser to apps; in France, 68 percent prefer the browser. In Spain, it's an even split, 42-42 percent, whereas in Poland, 45 percent prefer apps vs. 39 percent for browsers. An important datapoint is that in the UK, Spain, and Poland, users spend more time on the Internet outside the home than in, presaging the shift that many experts predict to a mobile-dominated online ecosystem. 4EIN World News Report, “Orange Mobile Exposure 2010 Study Reveals European Mobile Media Users Choose Browsers Over Apps for Mobile Internet,” PR Newswire, 10/6/2010
A 2010 report by mobile search company Taptu — which, it should be noted, is a browser-based service — crunches the numbers to show that, by sheer volume, the number of websites optimized for mobile touch far overshadows the number of available apps. Since that study early in 2010, however, the number of apps has continued to skyrocket. 5ReadWriteWeb, "Mobile App or Browser-Based Site? Report Says The Browser Will Win on Mobile," by Richard MacManus, 2/2/10
From a data traffic standpoint, it's a 50-50 split between native apps and the browser, according to a study by Zokem, a Finland-based mobile analytics company. This major study in 2009-10 of 10,000 smartphone users in 16 countries, covering 6.5 million usage sessions, found that browsers have a 54-46 percent edge in "face time," but the actual data consumed is at parity. However, the study also found that use of native applications is growing faster than mobile browsing overall. 6ReadWriteWeb, “Native Apps Account for Half of Mobile Internet Traffic,” by Sarah Perez, 9/20/10 Conclusion: inconclusive.
Is it an age thing?
Age appears to be one of the strongest correlating factors for mobile app or browser preference. A 2010 studay by Parks Associates shows a clear preference for apps in the under 35 set, especially in the key 25-34 demographic. As you go up in age, however, preference swings dramatically, in nearly a straight line. The older the user, the less preference for apps. That said, fully half of users across all ages are neutral on the question. 7ReadWriteWeb, “Consumers Under 35 Ditching Browser for Apps, Study Shows,” by Sarah Perez, 11/30/10
One part user need, one part business objective
Since the majority of users (except the older demos) seem as willing to use an app as a browser, depending on their need, perhaps the single biggest strategy driver should be how you can best meet that need: with an app, a browser, or both. Layer on top your business objective — are you trying to build a brand relationship, generate revenues, build readership, etc.? — and you will have two key data points to inform your mobile strategy.
Pros & Cons
Sooner or later, the ideal marketing strategy comes up against the often harsh realities of tactical execution. Mobile isn't easy. Unless resources and budget are abundant, trade-offs will have to be made. Here's a rundown of the pros and cons of each technology.
|Best control over UX||Fragmented device market: must be built separately for each platform; time- and resource-intensive|
|Can leverage full device functionality, i.e., location, camera, telephone, etc.||All platforms must be covered or major market segments will be missed|
|Typically loads and performs faster than current-state mobile Web||More difficult and expensive to maintain than similar functionality on mobile website; each app requires individual updates|
|Adds caché to brand||Requires mastery of platform-specific code and conventions|
|Can leverage push notifications||Cannot be discovered by search engines|
|Local content storage; can run without Internet connection||Requires device-specific programming skills|
|Potential monetization: app + in-app sales|
|Purchase and checkout process can be streamlined|
|Broadest audience reach: build once for most handhelds||Functionality more limited than apps|
|Can update advanced features using Flash or HTML5 once across all devices||HTML5 standards still not fully developed or implemented, Flash not fully supported across all smartphones|
|Can leverage location/GPS data||No push notifications|
|Can be discovered by search engines||
The Third Way: “Wapplications”
Is it an app? Or is it a website? An alternative to the either-or of app or website is a website that looks, feels, and acts like a native application. Some call them "wapplications" or "appsites." Take Google's HTML5 implementation for YouTube. It actually looks more like an app than Apple's own native YouTube app (the one you can't delete from your iPhone), and functions a lot more like YouTube's full website. (See the side-by-side comparison.)
ESPN is another great example. The mobile website and native iPhone app are practically identical twins, mimicking look, feel, and functionality.
A number of frameworks are available to simplify and streamline the "applification" of a mobile website, providing developers with tools to help create a native app look and feel, including Sencha Touch, SproutCore, YUI 3.2, jQuery Mobile, XUI, and iUI. 8Mashable, “HOW TO: Make Your Mobile Websites Act More Like Native Apps,” by Christina Warren, 8/18/10 And service platforms like Adobe Scene 7 automate the process of sizing, scaling, and delivering content in the proper format to a host of platforms, including mobile websites and native apps.
Mobile, yes, but how fast?
Whether your strategy includes apps, websites, or both, performance is a critical factor for user experience. Even as smartphone penetration and usage skyrockets, mobile performance remains problematic for many marketers. A recent Keynote Performance Index for mobile commerce showed that many top retail sites take more than five seconds to load their homepage even under the best network conditions, and that many have availability less than 99 percent. Keynote's overall Mobile Performance Index showed similar results.
The standard refrain has been to blame mobile network issues for performance problems. And while mobile data networks are still far from perfect, Keynote research indicates that server infrastructure, not the network, is becoming a bigger performance issue. Walmart's mobile site, for example, delivers consistently fast response; its pages are not the leanest of the sites studied, but it is using faster servers. That's part of the reason Walmart earned the top spot in Keynote and Yankee Group's "Best of the Anywhere Web 2010." 9Yankee Group Research, Inc. Anchor Report, “Best of the Anywhere Web 2010,” by Carl Howe, October 2010
Mobile networks are getting faster and faster, too. On Dec. 5, Verizon rolled out its 4G LTE network. PC Magazine's preliminary testing of the network achieved average speeds of 15Mbps, and a peak of 21Mbps. That's as fast as or faster than most wired broadband connections. Ironically, the magazine points out that, even at more modest speeds of 5.6Mbps, a user would blow through their $50 5GB data cap with Verizon in just two hours. 10PCmag.com, “Verizon LTE Blows Through Monthly Data Cap in 32 Minutes,” by Sascha Segan, 12/2/10
So the excuses for poor mobile performance are quickly running out. If Google can build a site that loads in 1.91 seconds, Strand Books in 2.40, Fandango in 3.03, and Walmart in 3.53, 11Keynote.com, “Keynote Mobile Performance Index - US, week starting 11/22/10”; “Mobile Commerce (Retail) Performance Index - US, week starting 11/22/10” then there's no rationalization for sites that take 5, 7, or 10 seconds to load.
How to build a fast mobile site
How the page is built is vital, but in mobile, overall page load is still a key factor, because even as they are improving, cell networks are inherently slower and inconsistent. Evaluating use cases and determining what is really important to users on the go should guide IA and content decisions. Limiting content to what's most important, and keeping pages small and downloads fast, will generally result in happier and more loyal users.
Validate, measure, monitor
Lack of performance monitoring technology is no longer an excuse for poor mobile performance, either. Tools now exist to validate, measure, and monitor mobile sites nearly as effectively as Websites. Keynote's Mobile Internet Testing Environment (MITE) 2, for example, includes content scoring against mobile best practices for 1,800 mobile devices, and recommendations for fixes. (You can get your free copy at http://mite.keynote.com.) And Keynote's on demand mobile cloud performance monitoring services also provide real-world data on site performance at the user/handset level (see: keynote.com/mobile).
Buying Advice: Start with a mobile website
Mobile is still evolving at a breakneck pace, and apps are and will remain a powerful phenomenon. But for broadest reach, biggest bang for the buck, most efficient rollout and maintenance, and searchable discovery, a mobile website is the logical starting point in a mobile strategy. Then, if and when budget and resources allow, individual apps for the major platforms serving your audience are great additions for enhanced user and brand experiences.