Can the new iPad deliver on its performance promise? Five things to consider before answering that question
7 Mar 2012
Speed matters, and the folks at Apple know this. The new iPad 3 is expected to deliver faster performance via a
(rumored) boosted Apple A5X mobile processor. As reported by Tech Crunch, “Apple states that the A5 SoC is ‘twice as fast’ as the Tegra 3 and the A5X offers ‘four times the performance.’”
But at the end of the day, it’s still a mobile device, hampered by many of the same constraints as smartphones, from battery power management to touchscreen lag to flaky browser cache. It’s also hampered by fast-growing user expectations. The average tablet user doesn’t think of their device as being a big-screen smartphone. Instead, they view it as a leaner, meaner laptop.
So the big question is: Out in the real world, can the new iPad truly deliver a satisfyingly fast user experience? It’s not a simple question to answer.
A few things to consider when you think about the iPad and performance:
1. iPad users are more similar to desktop users than they are to smartphone users.
In January, I dug into data spanning hundreds of millions of desktop and mobile transactions and found that iPad users are more similar to desktop users than they are to mobile users.
While iPad users view somewhat fewer pages per visit than desktop users (4.54 versus 5.14, 5.17, and 6.13), their average time on site and bounce rate were commensurate with the desktop crowd. Not a huge surprise. What’s interesting here is that, even though iPad performance lags behind desktop, iPad users seem willing to stick around for a longer desktop-like experience.
2. When sites are really slow, iPad users are as impatient as smartphone users. But when sites are really fast, iPad users are less impatient than smartphone users.
This is a really interesting dichotomy. Here’s an animated slide I created for Velocity EU last fall to illustrate this pattern in iPad users versus iPhone and Android users:
When things are really slow (as in page loads of 20+ seconds), iPad users bounce at about the same rate as Android and iPhone users. But as speed improves, iPad users tend to stay and their bounce rate gets dramatically lower — around 5% compared to 8% for iPhone users and 11% for Android users.
This is especially interesting given what we now know about conversion rates for iPad owners versus other mobile owners. (For example, over the Black Friday weekend, shoppers using iPads converted at a much higher rate than other mobile consumers, 4.6% vs. 2.8% for users of all other mobile devices.) Clearly, keeping your iPad traffic happy is a priority.
3. When it comes to generating page views, iPad trounces other mobile devices.
Again, no big surprise here. But it’s still interesting to see iPad’s growing dominance visualized, as in this animated graph depicting mobile traffic on a Strangeloop customer’s site over the course of two years:
You can see the seminal moment at the end of 2010 (right after Christmas, when it seemed like half the people in the western world suddenly had iPads) when iPad surged past iPhone and Android. While iPhone and Android made significant gains over 2011, page views via the iPad more than doubled — from less than 300K to more than 700K — in just one year.
You can see the same scenario play out over a much more compressed time frame in this next animation, which shows the adoption of an extranet application for another Strangeloop customer, a Fortune 500 company that had just launched a new internal app. Over just six weeks, you can see the mobile devices all starting out mostly neck and neck. Page views for the iPad slowly but surely lead the pack until week 5, when suddenly the iPad leaps ahead, leaving other devices in its dust.
In my opinion, this is just the beginning of the tablet’s dominance. It’s a no-brainer. The average person wants to see more content on a bigger screen.
But while people may embrace tablets, the tablet faces the same performance challenges as other mobile devices…
4. 3G performance is up to 10X slower than throttled broadband service.
As I mentioned above, most iPad users are browsing on their sofas. But many are not, and these folks are getting hit by brutal network slowdowns. Four months ago, I tried a quick-and-dirty approach to real-world mobile speed testing and found that latency can hit 350ms on a 3G network. It wasn’t pretty.
If you’re experiencing latency of up to 350ms on a page with 50 resources, that’s a whole lot of load time. Given the fact that iPad users (and tablet users in general) expect a desktop-like experience, this lag time is painfully unacceptable.
But network performance is just the first half of the problem…
5. 97% of mobile end-user response time happens at the front end.
About a year ago, I revisited Steve Souders’s four-year-old stat that says that about 80% of end-user response time occurs at the front end, and made a surprising discovery: while this number has remained pretty much the same for desktop response time, the front end is where a whopping 97% of mobile response time happens. What this means for tablet owners: When they aren’t struggling with network issues, they’re challenged even further at the browser level. It’s one uphill battle followed by another uphill battle.
Where does this leave site owners?
Tablets are one of the most exciting innovations to hit the market in the past ten years, and Apple is definitely leading the charge. But you can’t have innovation without a few headaches. Site owners are faced with the challenge of creating sites that work on three types of devices: desktop computer, tablet, and smartphone. These sites need to look good, load quickly, and contain enough functionality and content to satisfy demanding visitors. At the same time, they need to respect bandwidth limitations and bow to the fact that mobile network throttling is on the rise.
How are site owners going to do this? There’s no single answer. The solution lies in embracing a diversity of strategies, from responsive design to infrastructure investment to mobile-specific CDNs (to, of course, front-end optimization). It’s going to be an interesting few years.