iPad Usability: First Findings From User Testing

by Jakob Nielsen on May 10, 2010

Summary: iPad apps are inconsistent and have low feature discoverability, with frequent user errors due to accidental gestures. An overly strong print metaphor and weird interaction styles cause further usability problems.

"It looks like a giant iPhone," is the first thing users say when asked to test an iPad. (Their second comment? "Wow, it's heavy." )

But from an interaction design perspective, an iPad user interface shouldn't be a scaled-up iPhone UI.

Indeed, one finding from our study is that the tab bar at the bottom of the screen works much worse on iPad than on iPhone. On the small phone, users are likely to notice the muted icons at the bottom of the screen, even if their attention is on content in the middle of the screen. But the iPad's much bigger screen means that users are typically directing their gaze far from the tab bar and they ignore (and forget) those buttons.

Another big difference between iPad and iPhone is that regular websites work reasonably well on the big tablet. In our iPhone usability studies, users strongly prefer using apps to going on the Web. It's simply too painful to use most websites on the small screen. (Mobile-optimized sites alleviate this issue, but even they usually have worse usability than apps.)

The iPad's bigger screen offers reasonable usability for regular Web pages. Of course, there's still the "fat finger" problem common to all touch screens, which makes it hard for users to reliably hit small targets. The iPad has a read–tap asymmetry, where text big enough to read is too small to touch. Thus, we definitely recommend large touch zones on any Web page hoping to attract many iPad users.

Also, most Web pages offer a rich and overstuffed experience compared to the iPad's sparse and regulated environment; when an iPad app suddenly launches users onto the Web, the transition can be jarring.

For more than a decade, when we ask users for their first impression of (desktop) websites, the most frequently-used word has been "busy." In contrast, the first impression of many iPad apps is "beautiful." The change to a more soothing user experience is certainly welcome, especially for a device that may turn out to be more of a leisure computer than a business computer. Still, beauty shouldn't come at the cost of being able to actually use the apps to derive real benefits from their features and content.

First Studies

We conducted our initial usability studies of iPad apps and content a few weeks after Apple launched the device. We tested 7 users — all with at least 3 months' iPhone experience — but only one was an "experienced" iPad user.

(This user had only a week's experience — far less than the minimum of one year's experience that we usually request of usability study participants.)

Obviously, the findings from this research are only preliminary. However, we're releasing them anyway because the iPad platform is so different and is expected to attract considerable application development during the coming months. It would be a shame for all these apps to be designed without the benefit of the usability insights that do exist, despite the gaps in our current knowledge.

We tested the following applications and websites:

  • ABC player
  • Alice in Wonderland Lite
  • AP News
  • Art Authority
  • BBC News
  • Bloomberg
  • craigsphone (Craigslist)
  • eBay (both app and website)
  • The Elements (physics courseware)
  • Endless.com
  • Epicurious
  • ESPN Score Center
  • ESPN.com
  • Gap
  • Gilt
  • GQ magazine
  • GWR Lite (Guinness World Records)
  • iBook
  • IMDb (Internet Movie Database)
  • iverse Comics
  • Kayak (kayak.com)
  • Marvel Comics
  • MLB.com (Major League Baseball)
  • Nike.com
  • Now Playing
  • NPR (National Public Radio)
  • The New York Times Editors' Choice
  • Popular Science
  • Time Magazine
  • USA Today
  • virginamerica.com
  • whitehouse.gov
  • Wolfram Alpha
  • Yahoo! Entertainment

Wacky Interfaces

The first crop of iPad apps revived memories of Web designs from 1993, when Mosaic first introduced the image map that made it possible for any part of any picture to become a UI element. As a result, graphic designers went wild: anything they could draw could be a UI, whether it made sense or not.

It's the same with iPad apps: anything you can show and touch can be a UI on this device. There are no standards and no expectations.

Worse, there are often no perceived affordances (i.e., signifiers) for how various screen elements respond when touched. The prevailing aesthetic is very much that of flat images that fill the screen as if they were etched. There's no lighting model or pseudo-dimensionality to indicate raised or lowered visual elements that call out to be activated.

In contrast, long-standing GUI design guidelines for desktop user designs dictate that buttons look raised (and thus pressable) and that scrollbars and other interactive elements are visually distinct from the content.

The traditional GUI separation between "church and state" — that is, between content and features or commands — has carried over to modern Web design. Those 1993-style image maps are long gone from any site that hopes to do business on the Internet.

The iPad etched-screen aesthetic does look good. No visual distractions or nerdy buttons. The penalty for this beauty is the re-emergence of a usability problem we haven't seen since the mid-1990s: Users don't know where they can click.

For the last 15 years of Web usability research, the main problems have been that users don't know where to go or which option to choose — not that they don't even know which options exist. With iPad UIs, we're back to this square one.

Inconsistent Interaction Design

To exacerbate the problem, once they do figure out how something works, users can't transfer their skills from one app to the next. Each application has a completely different UI for similar features.

In different apps, touching a picture could produce any of the following 5 results:

  • Nothing happens
  • Enlarging the picture
  • Hyperlinking to a more detailed page about that item
  • Flipping the image to reveal additional pictures in the same place (metaphorically, these new pictures are "on the back side" of the original picture)
  • Popping up a set of navigation choices

The latter design was used by USA Today: Touching the newspaper's logo brought up a navigation menu listing the various sections. This was probably the most unexpected interaction we tested, and not one user discovered it.

Similarly, to continue reading once you hit the bottom of the screen might require any of 3 different gestures:

  • Scrolling down within a text field, while staying within the same page
    • For this gesture to work, you have to touch within the text field. However, text fields aren't demarcated on the screen, so you have to guess what text is scrollable.
  • Swiping left (which can sometimes take you to the next article instead of showing more of the current article)
    • This gesture doesn't work, however, if you happen to swipe within an area covered by an advertisement in The New York Times app
  • Swiping up

iPad UIs suffer under a triple threat that causes significant user confusion :

  • Low discoverability: The UI is mostly hidden within the etched-glass aesthetic without perceived affordances.
  • Low memorability: Gestures are inherently ephemeral and difficult to learn when they're not employed consistently across apps; wider reliance on generic commands would help.
  • Accidental activation: This occurs when users touch things by mistake or make a gesture that unexpectedly initiates a feature.

When you combine these three usability problems, the resulting user experience is frequently one of not knowing what happened or how to replicate a certain action to achieve the same result again. Worse yet, people don't know how to revert to the previous state because there's no consistent undo feature to provide an escape hatch like the Web's Back button.

Crushing Print Metaphor

Swiping for the next article is derived from a strong print metaphor in many content apps. In fact, this metaphor is so strong that you can't even tap a headline on the "cover" page to jump to the corresponding article. The iPad offers no homepages, even though users strongly desired homepage-like features in our testing. (They also often wanted search, which was typically not provided.)

In electronic media, the linear concept of "next article" makes little sense. People would rather choose for themselves where to go, selecting from a menu of related offerings.

A strategic issue for iPad user experience design is whether to emphasize user empowerment or author authority. Early designs err on the side of being too restrictive. Using the Web has given people an appreciation for freedom and control, and they're unlikely to happily revert to a linear experience.

Publishers hope that users will perceive content as more valuable if each publication is a stand-alone environment. Similarly, they hope for higher value-add if users spend more time with fewer publications rather than flit among a huge range of sites like they do on the Web.

Using the desktop Web, a user can easily visit 100 sites in a week, viewing only 1–3 pages on most of them. (For example, for one task in which B2B users visited 15 sites, they spent an average of 29 seconds per pageview.) Most sites are visited once-only, because users dredge them up in a search or stumble upon links from other sites or social media postings. Without real customer relationships, content sites have no value and 90% of the money created by users spending time online accrues to search engines.

The current design strategy of iPad apps definitely aims to create more immersive experiences, in the hope of inspiring deeper attachments to individual information sources. This cuts against the lesson of the Web, where diversity is strength and no site can hope to capture users' sole attention. Frequent user movements among websites has driven the imperative to conform with interface conventions and to create designs that people can use without any learning (or even much looking around). The iPad could be different if people end up getting just a few apps and sticking with them.

Card Sharks vs. Holy Scrollers

UI pioneer Jef Raskin once used the terms card sharks vs. holy scrollers to distinguish between two fundamentally different hypertext models:

  • Cards have a fixed-size presentation canvas. You can position your information within this two-dimensional space to your heart's content (allowing for beautiful layouts), but you can't make it any bigger. Users have to jump to a new card to get more info than will fit on a single card. HyperCard was the most famous example of this model.
  • Scrolls provide room for as much information as you want because the canvas can extend as far down as you please. Users have to jump less, but at the cost of less-fancy layout because the designer can't control what users are seeing at any given time.

The Web is firmly in holy-scroller camp, particularly these days: users scroll a fair amount and sometimes view information far down long pages. Even mobile-phone apps often rely on scrolling to present more than will fit on their tiny screens.

In contrast, card sharks dominate the early iPad designs. There's a bit of scrolling here and there, but most apps try to create a fixed layout for the pretty screen.

There's no real reason we can't have both design models: cards on the iPad and scrolls on the desktop (and phones somewhere in the middle). But it's also possible that we'll see more convergence and that the Web's interaction style will prove so powerful that users will demand it on the iPad as well.

Toward a Better iPad User Experience

Even our limited initial user studies provide directions for making iPad designs more usable:

  • Add dimensionality and better define individual interactive areas to increase discoverability through perceived affordances of what users can do where.
  • To achieve these interactive benefits, loosen up the etched-glass aesthetic. Going beyond the flatland of iPad's first-generation apps might create slightly less attractive screens, but designers can retain most of the good looks by making the GUI cues more subtle than the heavy-handed visuals used in the Macintosh-to-Windows-7 progression of GUI styles.
  • Abandon the hope of value-add through weirdness. Better to use consistent interaction techniques that empower users to focus on your content instead of wondering how to get it.
  • Support standard navigation, including a Back feature, search, clickable headlines, and a homepage for most apps.

Although our full report offers additional detailed advice, we obviously haven't yet developed a full list of design guidelines.

One big question will remain unanswered for a year or so until we see how daily use of the iPad evolves: Will people use the iPad mainly for more immersive experiences than the desktop and mobile Webs? In other words, will people primarily settle on a few sources and dig into them intensively, rather than move rapidly between many sources and give each cursory attention?

Maybe people will begin to use the desktop Web for more goal-driven activities, such as researching new issues or performing directed tasks like shopping and managing their investments. And they might use the iPad for more leisurely activities, such as keeping up with the news (whether "real" news or social network updates) and consuming entertainment-oriented content. We don't know yet. The answer to this question will determine how far iPad UIs have to move from their current wacky style.

Full Report (Plus Update)

The full report from the world's first iPad usability research is avauilable for free download. We've since completed newer user research with a later generation of iPad apps, and the report from that study is available by following the same link.

Share this article: Twitter | LinkedIn | Google+ | Email