The Complexity of Simplicity

By Luke Wroblewski

Published: December 4, 2006

“Simplicity doesn’t come easy.”

Though many business strategies and publications continue to trumpet the power of simplicity in the design of digital products, for lots of companies and product teams, simplicity doesn’t come easy.

“Making the simple complicated is commonplace; making the complicated simple, awesomely simple, that’s creativity.”—Charles Mingus

While there are many reasons why keeping things simple is difficult, I’ve encountered the following three causes quite frequently:

  • Perceived simplicity can often conflict with actual simplicity of usage.
  • Actions that provide real value—and drive revenue—often have formidable learning curves.
  • Gradual engagement, the most frequently cited solution for managing complexity, is actually quite difficult to design and build.

Perceived Versus Actual Simplicity

“A perception of complexity can turn customers, clients, or business stakeholders off before they ever actually use a product.”

Many of us carry a few preconceived notions about simplicity. We assume things that are easy to use don’t have a lot of options and, as a result, shouldn’t appear cluttered when we first encounter them. In the world of product design, this means plenty of whitespace, clear calls to action, and an overall reduction of content—in the form of visual elements such as type, images, lines, colors, shapes, and so on. When a product has these attributes, we are more likely to assume it is easy to use. It’s quite possible that it might not be, but the perception of simplicity is there.

Conversely, a perception of complexity can turn customers, clients, or business stakeholders off before they ever actually use a product. In a worst-case scenario, an evaluation based on an opinion that “this looks cluttered; therefore, it must be difficult to use” can prevent customers from ever even trying a product out. But as Don Norman recently suggested, an initial impression of complexity might actually be an artifact of a product’s simplicity. In “The Truth About Google’s So-called ‘Simplicity’,” he wrote:

“Why are Yahoo! and MSN such complex-looking places? Because their systems are easier to use. Not because they are complex, but because they simplify the life of their users by letting them see their choices on the home page: news, alternative searches, other items of interest.”

Edward Tufte, guru of information design, would likely agree with Norman. Tufte is known for his own criteria for measuring ease of use. He simply counts the number of links on a Web page. The more links, the better the design. His perception of simplicity is based on information density—how much screen real estate is devoted to useful information—in this case links—versus “chart junk”. In Figure 1, you can see this type of rich information in the first of two maps that represent the same geographical area with different levels of complexity. Though the second map, shown in Figure 2, conveys much less information, people are likely to consider it easier to use, because of its visual simplicity.

Figure 1—A detailed map provides rich information

Complex map

Figure 2—A visually simple map of the same geographical area conveys less information

Simple map

Cultural context can also sway people’s perceptions of simplicity. In many Asian countries, for example, congested spaces are quite common, and the benefits of collectivism, or integration with strong, cohesive groups, may outweigh those of individualism, or loose cultural ties. In this context, a lot of whitespace or an overall reduction in visual activity may be warning flags that a product lacks substance and thus requires significant investment with little payoff. In these cultures, information density is an indicator of activity and, therefore, an incentive for use. So dense visual information, like that shown in Figure 3, may be more attractive than the sparse layouts that invite exploration in the Western world, because of their perceived simplicity.

Figure 3—Visual density in physical and cyber space

Visual density in physical and cyber space

Regardless of the specific biases of individuals, notions of perceived complexity can prevent potential users from discovering the simplicity of a product’s actual use.

Creating Value

“The customers who create the most value both require additional features to support their high levels of use and are the most vocal about their needs.”

Prior to its being purchased by Google, YouTube counted 100 million downloads and 65,000 uploads per day. According to the Guardian, that’s “a creator to consumer ratio” of just 0.5%. The Guardian also noted that just 1.8% of all users write more than 70% of all Wikipedia articles. Similarly, the top 100 of digg’s 20 million users are responsible for 56% of the stories on the site’s home page. These statistics provide further verification of the Consumer/Synthesizer/Creator ratio that Yahoo!’s Bradley Horowitz outlined, in which 1-10% of a Web community’s members are responsible for a huge amount of its value.

The fact that less than 10% of a company’s customers are their primary content creators is particularly important when those users are creating revenue. Consider the case of eBay power sellers, who—though a small percentage of the company’s 200 million users—are responsible for a large part of corporate earnings. Some estimates even indicate that eBay does not monetize 90% of their users.

As any designer familiar with accommodating the needs of power users can attest, it is quite likely that the customers who create the most value both require additional features to support their high levels of use and are the most vocal about their needs. In situations where power participants create the majority of a service’s revenues, or value, companies are wise to listen and address their needs.

On YouTube, many of the features and content on the player page, shown in Figure 4, primarily benefit power participants rather than the majority of users who just want to watch videos.

Figure 4—YouTube player page

YouTube

Unfortunately the features and information that help power participants meet their goals can quickly complicate things for the average user. Of course, one could hide or gradually reveal advanced features, but at the cost of discouraging new users from adopting the behaviors that create the most value for the company.

Simplifying a product for the average user may penalize power participants—who are arguably a company’s most valuable customers—because the tools they need to successfully complete their tasks are out of direct sight or behind several layers of interaction.

Gradual Engagement Is Hard

“Gradual engagement … is often quite difficult to design and implement effectively.”

The most common solution for accommodating the needs of both average and power users without negatively impacting the effectiveness of either group is to gradually reveal complexity as users let you know they are ready for it. Though clearly a noble pursuit, gradual engagement of this sort is often quite difficult to design and implement effectively, as it is likely to require complex behind-the-scenes considerations for a product development team.

In order to present the right user interface to the right user at the right time, designers must track multiple types of users and their various states, then map those contexts to an appropriate presentation of features and content. This, of course, is a non-trivial challenge and, as a result, is often done poorly. If the logic is not thought through completely, some users end up with too many options, while others feel they have too few.

As an example, consider the way Microsoft Office 2003 menus display only the items users use most frequently while hiding the others, as shown in Figure 5. This gradual engagement feature often makes it seem like the option you need—and are sure you have used before—is unavailable, when actually it is temporarily hidden on the very menu you are looking at.

Figure 5—Menus in Microsoft Office 2003

Office menus

An example where gradual engagement arguably works better is the search experience on Google. Though often cited as an example of a simple design, Google search was actually built for expert users. According to Product VP Marissa Mayer:

“Novice users will enter ‘tell me when it will snow in NY today’ and get no valuable results. Soon thereafter, they will end up typing ‘weather new york’ and see that the results are more valuable. Voila! An expert user. The learning curve in search is steep, but quick.”

Enabling this experience, however, requires all the computational power that an engineering powerhouse like Google can muster. Not all companies have such capabilities.

A much easier and, therefore, more common design and development solution is to simply expose all possible options to all users and have them select what is most appropriate for their particular needs. In this case, the design team doesn’t make decisions for users that may not be appropriate to certain contexts. This approach, however, doesn’t do much to simplify a product’s design. In fact, it probably has the opposite effect.

Despite all of this, we should not think of the problem of perceived versus actual simplicity, the need to enable power participants to create value, or the difficulty of implementing effective gradual engagement as reasons to give up on the pursuit of simplicity. Being aware of these considerations is actually likely to make our jobs simpler.

13 Comments

Fantastic article. I’ve been thinking about the concept that you refer to as gradual engagement—in particular with regard to Mac OS X, as an example.

The advanced functions in Mac OS X kept revealing themselves to me at a rate that made it one of the most interesting experiences. However, some of my peers don’t seem to discover many of these hidden powers at all. I was inspired to understand how Apple did that and how they could make it more universally revealing.

Do you know more about how gradual engagement is designed and what kind of people design it?

Of course, it makes sense to make expert features available to experts. On the other hand, experts do not necessarily prefer more feature-rich products—for example, see Thompson, Viana, and Rust, 2005. So what gives? I think the answer is that, while experts do use expert features, even they rarely use them. Most of the time expert features are a distraction even for experts. That doesn’t necessarily mean designers must eliminate or hide expert features, but it does suggest that designs should be proportional, with commonly used features easy to see and select, and rarely used expert features being less obtrusive, even if less convenient. For example, most links might be on the same page, but some links might be larger and bolder than others. Some links might be available only via a drop-down menu—things like that. By muting the expert actions to allow faster completion of the most frequent actions by both experts and novices, we can optimize the aggregate performance of all users.

Interesting topic and a nicely written article.

Simplicity and complexity are two sides of a coin, and they have been a topic of interest in various design fields. I did a literature review on engineering design complexity when I was in graduate school a few years back. There were various definitions of design complexity. One of the most common approaches was the use of information theory. One that interested me particularly was Nam Suh’s (see book).

A system could be complex in terms of what it is designed to achieve; that is, the design problem. It could also be complex in terms of how it is designed to be used; that is, the design solution. A colleague of mine once told me that the elegance of a design could be conceptualized as the complexity of the design problem divided by the complexity of the design solution. I thought it was very nice way of putting it.

Great piece! I think simplicity itself is a perception of the user. A product can be simple only if it plays to that perception and only for that user, which gets complex for multiple users. I like to look at a three-button mouse as an example. It works for a right-handed person, but can be configured for a left-handed person, too. I have put down my thoughts on my blog.

Well, unfortunately, users do not experience an interface in aggregate; they experience it individually. When one person is frustrated, he or she gains nothing from knowing that most other people aren’t!

Regarding Office 2003’s gradual reveals, they have admitted that it didn’t really work as well as they’d hoped: 90% of new feature requests for Office Vista were for features already available in 2003! (see Jensen Harris, MS Office Vista user experience team manager’s blog).

A better practice is to at least allow power users to migrate features to more accessible places should they choose to. One feature of Office I’m surprised more people don’t use is the ability to create your own toolbars. I have a single, tiny floating palette in each Office application that contains the few—about 20 each—buttons I ever click. I use shortcuts for all the general stuff like Bold or Save. This allowed me to migrate deep expert features I use to the surface at my choosing, while removing from view the generic features that I shortcut or never use.

Eric,

Gradual engagement is also known as progressive disclosure, and Jakob Nielsen recently wrote up an interesting post about it:

http://www.useit.com/alertbox/progressive-disclosure.html

Michael,

The distinction in the cases I’m talking about is between power users and power participants. Power participation refers to the acts that drive revenue and value for a business. Very few people do them, but they are what keep a company afloat: uploads for YouTube, news story submissions on digg. So even though a very small portion of users employ the features that enable these behaviors, they need to be exposed/communicated to all, because the value of their adoption is so high.

Shaun,

“The elegance of a design could be conceptualized as the complexity of the design problem divided by the complexity of the design solution” = interesting!

LeMel,

I agree with you about the Microsoft Office toolbars. I’m using those as an example of gradual engagement done wrong and evidence of the fact that progressive disclosure is hard to do right. Also, most people do not customize. Even sites/portals that are designed strictly for customization have 75% of people accepting the defaults. That’s why smart defaults are so important:

http://www.lukew.com/ff/entry.asp?419

Great article! The simplicity versus power/features dilemma seems to be in the air. Have you seen Don Norman’s recent article? http://www.jnd.org/dn.mss/simplicityishighly.html It makes a refreshing read. Even if product designers manage to reconcile simplicity and feature richness, some customers may still prefer complex-looking products!

The MSDN site actually published an article, “Powerful and Simple,” in the context of the Vista user experience design and guidelines. Some interesting definitions of powerful, the idea of simplicity versus simplistic, and practical approaches to gradual engagement can be applied to Web sites and Web applications, too.

Powerful and Simple

If you transfer the discussion to product design, I think that, when simplicity is based on cultural and humanistic values, it often outperforms complexity, just because it is understandable.

I’ve seen a similar statistic on WalkJogRun.net. There were 30,000 routes created over the last year between 278,000 visitors. That means only 10% of the visitors worked out how to add a route or at least wanted to. Humph. We are working on redesigning the menu structure to try and move some of the more advanced features like Google Earth export somewhere they won’t confuse regular users. Great article and certainly food for thought in light of our goals for the site—to make it accessible enough for everyone to add a route.

The simplicity versus complexity debate rages on, because the topic strikes a nerve for many usability and design experts with the following question: Does simplicity sell? This question, having come from a leading practitioner, Don Norman, hits a bit too close to home for comfort. Design experts have been preaching the value of simplicity, user friendliness, etcetera.

I simply appreciate your simple explanation of something not that simple. I am writing a movie about simplicity. Not simple at all. Could you tell me what you would like to see in a movie about simplicity?

I really enjoyed reading this post; lots to glean from it! One way I have simplified my life is that I don’t have a Facebook account. :)

Join the Discussion

Asterisks (*) indicate required information.