Persuasive Design: New Captology Book

by Jakob Nielsen on March 3, 2003

Summary: Review of B.J. Fogg's new Persuasive Technology book, which provides useful principles on how to think about creating persuasive design, but rarely gives detailed design guidelines. The exception is a section on enhancing website credibility.


It is a rare book that defines a new discipline or fundamentally changes how we think about technology and our jobs. Dr. B.J. Fogg's new book, Persuasive Technology: Using Computers to Change What We Think and Do, does all of this. I highly recommend that you read it for two reasons:

  • The book's indispensable design advice will grow your business.
  • You must teach your children to recognize this new class of manipulation.

Human Factors Changes Slowly

Human factors is a very established profession with a well-documented foundation that remains the same decade after decade. Once you have learned the profession’s basic principles and practical skills, you can apply them to any new technology that comes along. Human factors is concerned with human behavior , and homo sapiens doesn't have a Moore's Law to change its capabilities every 18 months.

Because of the robustness of human factors principles, many of the best books in the field were written in the late 1980s and early 1990s -- a period in which usability researchers crystallized the formerly academic discipline and made it practical. The two big stories were:

  • 1975-1983: following an early emphasis on military design problems, human factors' primary focus shifts to human-computer interaction .
  • 1989-1994: discount usability engineering is victorious over earlier cumbersome methodologies, and usability inspection methods become codified elements of the usability toolkit.

The growth of Web usability from 1994 to 2000 had huge practical impact, but didn't involve much new theory. We employed the methods we already knew and tested websites to document the many stupidities that dominated the early days of "killer design."

Persuasive Interaction

After ten dark years of fighting (and partly conquering) user-hostile design without much theoretical progress in HCI, Dr. Fogg has now opened the field's next frontier with his work on "captology" -- computers as persuasive technologies.

Persuasion in itself is obviously not new. From Cicero's oratory to modern TV commercials, communicators have tried to persuade audiences. What's different is that websites and other computerized designs are going beyond one-way rhetoric and becoming interactive. Doing something is much more engaging and thus potentially more compelling and persuasive than passively receiving messages.

Dr. Fogg has researched captology for many years and presents many interesting examples of persuasive design, both from websites and from information appliances. The book also includes many principles of persuasion, which are useful as a systematic way of thinking about different ways to make your own designs more persuasive. Mostly, these principles are not detailed enough to serve as prescriptive design guidelines . Also, not all the principles will apply to all design problems. The slightly vague nature of the book's advice is inevitable given that the captology field is still in the early stage of development.

As an example, one of the book's principles is that of tailoring information so that it's more specific to the individual user's circumstances. A website to promote tax cuts, for example, could calculate the amount of money a user would gain from a proposed new tax law. This is more persuasive than talking about tax reductions in general terms, partly because specifics beat generics, and partly because it personally involves the user, who must enter data to make it work.

Similarly, a financial website could use the principle of simulating cause and effect to encourage users to save more for retirement. Its simulations could be both symbolic and realistic. For example, it could show both a curve of the user's growing nest egg and a photo of "the hotel you can afford to stay at when traveling to Hawaii after you retire" for different levels of monthly investment.

Web Credibility: The Most Useful Part of the Book

The book's most immediately useful part is the section on credibility, which includes a rank-ordered list of how fifty-one design elements help or hurt a website's credibility. The four most harmful elements are:

  • Links that don't work (-1.3 on a 7-point credibility scale)
  • Content that is rarely updated (-1.7)
  • Links to sites that lack credibility (-1.8)
  • Ads that are indistinguishable from content (-2.1)

Note that poor linking accounts for two of the four most harmful design elements. Most companies probably don't even have a formal linking strategy, but links are the Web's foundation, and they're well worth some consideration. Good links, both incoming and outgoing, boost credibility almost as much as poor links lower it.

Many of Dr. Fogg's findings have been confirmed by independent research, making them very credible. For example, the importance of links was stressed in my own studies of how people read on the Web.

Also, one of Fogg's main credibility guidelines is the principle of real-world feel : a website's credibility increases when it shows the people or organization behind the site. This principle was also found in our usability testing of e-commerce sites. For example, users were more likely to buy from a coffee site that included a photo of the company's coffee roaster. Seeing the equipment increased people's confidence that the company actually had coffee on hand and would send their package if they placed an order.

Web persuasion is where the action is today, and persuasive design will have an even bigger impact when it becomes more tightly integrated with our environment. This is a clear trend as we abandon the PC as the center of interaction and the key viewport on the Net.

Fogg emphasizes the potential of captology for mobile design, which can leverage the kairos principle of offering suggestions at opportune moments . Other situated designs can benefit as well. For example, an interactive restaurant menu could replace the little heart icons that symbolize healthy meals with personalized icons to highlight dishes that agree with an individual patron's diet, whether it's Atkins, vegan, or the need to avoid nuts (except for those three kinds that are okay). No matter how complex a person's food preferences, computers can handle the problem and thus encourage diners to order more adventurously instead of sticking to the few items they know.

Is Persuasive Technology Ethical?

To some extent, this is an irrelevant question. Now that the book has been published, there is no doubt that companies will embrace the guidelines and implement them on websites and in other interactive technology. It will happen -- which is why I believe that one of the main benefits of Fogg's new book is that it can help concerned parents recognize interactive persuasion and teach their kids how to deal with it. We have long known about the persuasive capabilities of non-interactive propaganda movies and TV commercials. And, as with other media, it won't matter if parents like interactive persuasion or not: it will be there, and they'll have to deal with it.

One type of persuasive technology is clearly ethical: when users ask to be persuaded. In many cases, you know something is good for you but you may not have the willpower to meet your goals. Interactive systems could track your progress and motivate you to do better.

A current example is the FitLinxx website, which is networked with the exercise equipment in many health clubs and can track how much weight you lift and how many calories you burn at each workout. In addition to being a good example of integrating the physical world with the online world, FitLinxx also exemplifies positive persuasive technology: The simple act of tracking your progress and comparing the statistics with your goals can motivate you to exercise more.

Similarly, a Harry Potter-like talking refrigerator embodied with computational initiative (and networked with your health club) might say, for example, "That's the third snack today; you didn't exercise enough yesterday to eat this much." That nagging voice (at exactly the opportune moment) could help you stick to your diet.

The Ethical Gray Zone

Other uses of persuasive technology, where people with good intentions make decisions for others, are less clear-cut, but probably ethical in many cases. For example, parents or doctors could give diabetic children devices to monitor their food intake and motivate them to eat in ways that would minimize their condition's negative impact.

Helping sick kids improve their health is surely ethical, but what about the equally well-intentioned parents who give their children talking plush dinosaurs that indoctrinate them into the parents' ideology? "Clean up your room" or "It's fun to play together" might be acceptable, but what about certain judgmental attitudes?

Persuasion is clearly unethical if it relies on deception. A current example from the Web is banner ads that try to persuade users to visit certain websites by masquerading as dialog boxes. These ads encourage a certain behavior, not by trying to convince the user that the behavior is a good idea, but by persuading users that they're doing something completely different than what they're actually doing.

Ethical if you asked for it; unethical if it's deceptive. That leaves the vast majority of persuasive design in the gray zone. It's hard to judge the ethics of persuasion based on the values embedded in the design. If you share those values, you will approve. If you disagree, you will disapprove.

Dr. Fogg's chapter on ethics offers a deeper analysis than the few issues I have raised here, but I think he mainly leaves the question unresolved.

Usability usually doesn't involve many ethical dilemmas. Making a website easier to use benefits both the company and its customers, so there is not much to worry about in traditional usability projects. Persuasive design clearly is more complex, and we probably need a few more books and many more case studies before the ethics become clear. For now, it's clearly an issue you need to pay attention to.


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