When you stop to think about it, the job of interaction designers is about persuading people to do something. We design juicy-looking buttons and place them “just so” on the page to entice people to click them. We remove distractions and streamline processes so that customers glide effortlessly from browsing to buying. We create whole applications aimed at getting people to check in more frequently, diet more effectively, work out more consistently, or do more, more, more of whatever it is that meets our persuasive goal.
What Is Evil Design?
Sometimes our aim is charitable: we are persuading people to do something that benefits society more than it benefits them. Sometimes our aim is motivational: we are persuading people to do something that will benefit them even if they wouldn’t choose to do it unaided. Often our aim is commercial: we are persuading people to trade something that will provide equitable benefit to our company and to them. Sometimes however, it seems aims are less virtuous, and companies seek to get customers emotionally involved in doing something inequitable—something that benefits the company more than it benefits the individual.
And that, perhaps, is the definition of evil design: to get customers emotionally involved in doing something that benefits you more than it does them. Now, your first reaction may be to deny that this would ever happen in your company, much less that you’d be complicit in it. But somewhere, someone in the interaction design profession is building these interfaces, because they aren’t accidental. They are the result of applying psychological principles or design patterns to arrive at an emotionally engaging experience.
Example Evil Design Patterns
Let Users Advertise Their Status
It’s hard to define which interfaces are truly evil because different people will derive different levels of intangible value from their interaction with a site or application. Take for example the ubiquitous email signature “Sent from my iPhone” (see Figure 1). I hope that the person who created that default signature setting got well rewarded for their work. It is the perfect embodiment of effortlessly viral aspirational content. Taken at face value, it is little more than an advertisement for a product—something that benefits the company more than the customer. Yet users seem strangely reluctant to change it. This inertia stems at least in part from what that small phrase says about them as individuals. Even more than half a decade after the device’s release, the phrase is still prevalent at the bottom of people’s emails, and it can’t be because no one knows how to change the setting.
The iPhone default signature remains partly because it’s boastful in a socially acceptable way. The designers found a way to let users advertise their status as owners of a shiny, desirable gadget. This design pattern, encouraging users to build and advertise their status within a community, lets people feel more important. It affects customers at an emotional level but serves a very financial goal for the company. Who benefits more? Financially, it’s quite clear. Emotionally, it’s hard to say.
Pre-Pick Your Preferred Option
Psychologists have known for a long time that if they show you specific words or pictures beforehand, you’ll find it easier to recall those items, or related items, in a later test, even after you have consciously forgotten the specific words. This effect is called priming.
This is one reason why you see so much brand-related advertising. Rather than specifically telling you to buy a certain thing, the advertisers build a picture of a brand associated with a specific location (a bar), emotion (happiness), or occasion (a celebration). When you are in a bar for someone’s birthday party, all that brand priming comes straight back from your subconscious brain and hits you. You aren’t going to get a rum and cola; you’re going to get a Bacardi and Coke. You’re not going to buy a beer; you’re going to buy a Budweiser.
Couple this priming with inertia and you have the perfect conditions for encouraging people to select a given option. By making the option seem familiar beforehand, and then presenting it as the default choice, it’s easy to guide people’s behavior.
Use Negative Options
In the United States, many employers offer personal retirement savings plans, 401(k) plans, for their employees. Sign-up is typically optional and, once started, employees can choose how much to contribute. In their 2001 Quarterly Journal of Economics article “The Power of Suggestion: Inertia in 401(k) Participation and Savings Behavior,” Brigitte Madrian and Dennis Shea reported that those employers who automatically enrolled new employees saw higher rates of contribution than when employees had to make the personal effort to enroll. In both instances, it turned out that many employees couldn’t be bothered to change the default (pre-picked) state.
This inertia could have major implications for the future financial health of the country since participation in personal retirement plans reduces the burden on future government-run social services. Interestingly, many of the automatically enrolled employees kept both the default contribution level and fund allocation, which shows that it makes good sense to set the default option to the behavior that you want to encourage.
The last example almost certainly falls into the category of motivational rather than evil design. But the designer of that company’s healthcare policy leveraged the same concepts of priming and inertia that are also used to trick us into signing up for junk email. However good we get at spotting the marketing opt-in checkboxes on registration forms, we must still read them very carefully. Is the box pre-checked? Is it offering to opt us in or to opt us out? Is the second box, which talks about sharing details with third parties, also pre-checked? Is that one opt-in or opt-out? The company’s goal is to get you to start receiving its emails because you’re less likely to make the effort to unsubscribe after you’re signed up. This is the basis of the negative options design pattern: sign people up to receive something until they specifically choose to stop receiving it.
Take for example the three innocuous-looking checkboxes on the Hotel Chocolat website that can have major repercussions on the future size of your email inbox and those of the people you send gifts to (see Figure 2). Read carefully: which boxes need to be checked or unchecked to not receive emails?
A study by Steven Bellman and his colleagues, described in the 2001 ACM article “On Site: To Opt In or Opt Out? It Depends on the Question,” found that simply using a negative option doubled opt-in rates. Phrasing an option as “Notify me about more surveys” with the checkbox unchecked (that is, people must take an action to get the additional surveys) led only to a 48 percent uptake. Changing the text to “Do NOT notify me about any more surveys” while still leaving the checkbox unchecked (that is, people need not take any action to get the additional surveys) produced an astounding 96 percent uptake. Of course, as Bellman notes, not all of the individuals who opted in (or rather didn’t not opt in) in the second example are actually good prospects, but the apparent consent rate is incredibly impressive. Often it’s easier just to keep doing what we’ve always done than to make a change, even if that change would save us time, money, or the hassle of a spam-filled inbox.
Make Options Hard to Find and Understand
It took four months and 3,000 downloads for someone to claim the reward that PC Pitstop hid in the Terms of Service document for one of their products. Only 0.06 percent of ISP Embarq’s customers read the updated Terms of Service document that allowed the company to spy on, and insert ads into, each data packet they sent through their Internet connection. Putting opt-outs in out-of-the-way places and obfuscating with hard-to-parse text seems to be a common business practice for introducing changes that are more likely to benefit the company than the customer.
I’m not approaching this topic from some holier-than-thou angle. I’m sure people can poke many holes in the persuasive techniques I’ve used in the products I’ve helped to design. Instead, I’m suggesting that a catalog of these techniques will help us to recognize and respond to them. Sometimes that response might be to avoid giving our business to sites that we see using the pattern. Sometimes the response might be to repurpose a pattern for good. Sometimes the response will undoubtedly be “that’s exactly what I was looking to implement in my product.”
My response has been to search out these persuasive techniques and describe them as design patterns. This article touches on four of the fifty-seven patterns I describe in detail in the book Evil by Design. As Don Norman says in the book’s foreword, “the more the tactics are understood, the more readily they can be identified and resisted, fought against, and defeated.” Please join the conversation about these techniques online at evilbydesign.info.
Comments are closed.