Are We The Puppet Masters?

About IxD and ethics.

Through the designs we create, we have the ability to directly influence another person’s behavior. The ethical implications of this are important and not easily definable. I was interested in ethics before I ever considered becoming a designer, but the lessons I learned while studying philosophy impacts the way I view my designs. In nature, our goal is a good one. We strive to help others by improving the interactions that define their life. This drives us to create and innovate new ways of interacting with old concepts. The question remains, do we have the right to influence another person? Further, are there guiding principles we can follow that can keep us on the moral path? The answers to these questions rests on the shoulders of the whole community, not a single person or group.

History Lesson

Christopher Fahey’s blog post ‘Who Watches the Watchmen?‘ was the catalyst that lead me to exploring this topic. What struck me the most about Chris’ post was how the design of the watchclock dehumanized the role of the guard. It turned the guard into just a part of a system by requiring them to visit key locations during their patrol in a set, unchanging, order. Immanuel Kant wrote ‘it is always wrong to treat other human beings as mere phenomena, as objects of your own pleasure or pain. Rather, human beings must always be treated as of infinite worth…’, which speaks to the heart of this issue. It is up to us as designers to ensure that people don’t simply become actors of a system, but real people that have desires, needs, and goals. My response to Chris’ post started on my blog, and eventually moved on to the IxDA Discussion List.

Dangers of Puppet Masters

User Experience, and all of its various children, come from an age when programmers and engineers just built things. These manufactured interactions provided little meaning to end users and caused much pain and frustration. It’s easy as designers to fall into a similar trap of just designing for design sake. Software and hardware become more like art, being an expression of the designer rather than a tool for people. Only these pieces of art can lead people down undesirable paths they neither want or need. This is the real danger of being a Puppet Master.


Cell phones today are measured by the level of interactivity, not by its ability to make phone calls.

Cell phones today are measured by the level of interactivity, not by its ability to make phone calls.

The watchclock, which inspired this investigation, is a system that has direct influence on a person’s behavior. It was built to ensure that security guards were doing their job. What was missing from the design was a consideration for the behavioral need of the security guard to mix things up. The design of an interactive product is an exercise in evaluating current behavior, and putting something in place that supports that behavior or alters it in a more meaningful way.

But, how do we know if what we are designing is ‘good’? Regulations like the US Section 508, a set of policies that any U.S. government website must follow to accommodate web users with disabilities, provides some guidance for web designers. Policies like this have a direct influence on designing interactions for the web, and provide a guiding light for those not sure if they are doing the right thing in terms of accessibility. One unavoidable aspect of interactive products is the simple fact that people do interact with them. By focusing on all the possible good things an interactive product can be capable of, the chances of designing something ‘bad’ can easily be avoided.

Designer Point of View

Only a designer can go into great detail why machines like this are horrible.

Only a designer can go into great detail why machines like this are horrible.

Designers are gifted with a certain perspective of the world that can cause much frustration and wonderment. The average person doesn’t have the filters in place to see when they have been ignored by the product they are using. Occasionally, people can tell when something wasn’t designed, but they normally deal with the damages done physically, mentally, and socially. In an attempt to stop the pain, designers create interactions that look to discourage undesirable behavior and promote desirable behavior.

Using our unique perspective, we can learn a lot about what makes a group of people tick. The lessons we learn from them gives us insight into what types of interactions they are really looking for, and how their lives might be improved. Though people have a hard time knowing when they have been ignored by a company, there is instant recognition when they know their voice has been heard. You begin to hear statements like “I love this …” or “I will never use one of those old …. again.” Getting people engaged at this emotional level, and harboring trust between the business and the consumer is a powerful persuasion tool to break undiscovered territory.

Understanding the importance of choice is an attribute of being a great designer. In order to do ‘right’ by the people interacting with our work, they need to know what their choices are, and how those choices will impact them. Making an interaction a success means providing people with the choices for their situations. Providing transparency on what options are available, we allow a person to be free to make up their own mind. People don’t want their choices to be made for them, as this makes them feel powerless and dehumanized.


An argument can be made that Interaction Design should be called Influence Design. The risk with having lots of influence over an interaction though is it can easily lead to control. We may try to trick the user here and there, but in the end we have to leave the choice in their hands. In a good design, these choices are invisible to the user, but still easily understood. They are road signs we place as someone becomes engaged with the interaction. Clearly defined ‘Exit’ signs are key to giving people the freedom to interact, but not fall under the control of the user.

José Antonio Martínez-Salmerón of frog describes it best in his article ‘The Power of Persuasion’. “No matter how many choices a consumer is faced with, a product that’s relevant, safe, and personally fulfilling will always stand out.” The number of choices available to people are greater than ever today. The challenge is making those choices meaningful and relevant for their life. In the end, a person has to be free to go against what we have laid out for them, even if it means they will use a different product. We are not their keepers, but merely stewards that look to improve the human situation.

We are not their keepers, but merely stewards that look to improve the human situation.

In order to find some ethical guidance around the amount and type of influence designers have, we need only look to the Poka Yoke Principle. We attempt to steer people away from possible accidents, or damage, and lead them towards meaningful experiences. This is a great way to build trust in the people you serve, and trust is the best type of loyalty you can have with a customer because they know you are looking out for them. Having built up this trust, people are more inclined to abandon their bad behaviors, and take up new ones.


It appears that we have painted ourselves into a corner. Having direct control over another person’s behavior is wrong, as they did with the watchclock. But, we can’t help influencing a person’s behavior with the interactions we design. Reviews of the interactions we create should ensure that choices are not limited or completely taken out of the hands of people. Being a designer is a true calling, one which drives us to make the world a better place, a more user friendly place. We can’t become overprotective, but rather acknowledge the importance of someone making a mistake, or even completely failing. People learn from their failures, and if we are watching closely we can learn from them too. These lessons help us create and innovate to fix problems of yesterday, and possibly help expose the problems of tomorrow.

Brad Nunnally

Brad Nunnally is a User Experience Design Consultant at Perficient based in St. Louis, MO. Aside from writing, plotting UX world domination, and tweeting a whole bunch , he fills his time playing with his son and dog.

16 comments on this article

  1. Brad,
    I think for a limited scale of solution types what you are suggesting is true. However, if we are to create organizational & global cultural change we have to think bigger about our role as designers. Service designers work at the level of policy decision making and IxD’s work within the touch-points that enable policies have to influence, persuade and even insist on specific micro & macro behaviors.

    I also don’t really understand these “new” arguments. We KNOW that advertising has been pulling our cognitive and emotional strings for the last 100 years in turn creating an entire culture for which we’ve allowed ourselves to be more than influenced, but totally manipulated. So we know Design DOES do this.

    To me design is the tool. the ethics is not applied to the tool, but rather the outcomes and unfortunately the ethics like in all cases are arguable, relative and totally subjective, just like in policy making itself.

    In fact, I would argue that pursuing the role of puppet-master would strengthen our position in effecting greater change and having an even more positive effect on the human condition.

  2. Joe R on

    Nice article! Seems like a lot of blah blah blah. Ethics in designing that’s an interesting concept. Unfortunately, money is the driving force, it’s about profit and how to get the consumer to purchase your product or better yet, your company’s product. That throws ethic right the window, with the watchman’s watch! Good article, though! Nice work!

  3. Laura on

    Driving purely for profit is an ethos, too.

    There’s no doubt in my mind that design affects users in ways that some might want to measure in moral terms. But I think it’s a subtler thing than the watchclock. Often in design we deliberately restrict options and flexibility. It’s not always dehumanizing. It can be empowering.

    I wonder more how the patterns affect us as designers. How much do we do X because it’s easy and expected, even if it might not be the best choice? To take an analogy, how much of business planning is funneled into what fits conveniently into a powerpoint presentation? Do we become hammer wielders instead of chisel users because we have hammers readily available? Maybe I’m getting to abstracted here.

    In the end, who’s in charge? The designers? Or the people who choose to use — or not use — the design product?

  4. Brad,

    This is an interesting and recurring debate. Couple things that spring to my mind are

    a) users own the experience (even if design shapes or constrains it)

    b) ethical matters involve human outcomes and to me would relate most to your concerns around usability. artistry, craft, and design skill, method, and process to me are not ethical per se. A good gun design — there is such a thing but the ethical question concerns the thief, prison guard, gun dealer, bullet manufacturer, as well as gun maker…

    c) good and bad are attributes of use, and thus of utility, and thus of context, for context informs uses. (at whom is that gun pointed??).. again, it seems that context can so exceed the design intentions and choices behind the thing that the designer need hardly worry about ethics

    d) ethics unites causes and effects, or intentions and consequences, and so it seems that a designer’s intentions are only part of the context in which consequences play out (is that å thief on the run?). “Social interaction design” is not social engineering, as experience design is not experiential control. I’m not comfortable with the suggestion that design causes behavior, or so constrains behavior that the consumer can do only what we had in mind for her to do…

    e) I believe the questions regarding the consequences of design may have much in common with the marketing and advertising world in which our products test their fates. And as much as marketers think they know us, and as much as advertisers believe they manipulate us, they don’t, and they know it. So I don’t worry too much that design exacts influence on the end-user

    The user is always a step ahead of the obvious. To quote the structuralists, “a system is defined by that which escapes it.” Or to quote comedian Mitch Hedberg, “If i ever have to take another test I’m going to take it in a restaurant. Because the customer is always right.”

    I suspect this topic recurs because there’s something to it. Of which both your post and these comments must be proof!

  5. Mike on

    @adrian, wouldn’t the Panopticon be a counterexample? I’m not sure if you intended this, but your argument sounds almost libertarian. If design can’t cause behavior, there are no ethical problems for designers; guns don’t kill people, people kill people, etc.

  6. Brian on


    Nice article, UX (and all it’s subsequent derivatives) designers are stewards of the user. That doesn’t feel puppet master like to me and I’m not worried about designers having influence over me. Maybe I’m naive.

    Simply by having a profession for this, we’ve come a long way.

    – Early factory line machines used to chop peoples hands off on a frequent basis.
    – Early monochrome monitors used colors tough on peoples eyes.
    – Early software required hundreds of clicks (or typing) for tasks that are now take a few clicks.

    These are all instances where “designers” didn’t know any better, and now they do. I don’t think they intended to cause harm or be a puppet master.

    Now there is a profession dedicated (or several really, depending on what people are designing for) to making objects / software / machines easier for humans to use and better.

    Factory robots now assist humans in manual labor tasks better than just using humans (or at least more efficiently / safely). UxD folks make websites / software / systems easier to navigate and use often based on user studies / feedback.

    Yes, some of it is designed to sell more, influence people to vote a certain way and change people’s buying habits or behaviors, but the ethical question really comes to what people are designing a product for, not the profession itself.

    Using Mike’s example, if you are designer who makes a gun work more safely (like a better child-proof safety) but this causes the gun to sell better, are you at fault for people who kill with this gun? If you feel so, go work for a company who doesn’t produce lethal items. Some people would take solace that you designed a better safety.

    I guess when I boil it down, I don’t think the ethical question is specific to user designers, because they influence behavior. I think the fact that people out there looking out for users is a good thing.

  7. @Dave – “In fact, I would argue that pursuing the role of puppet-master would strengthen our position in effecting greater change and having an even more positive effect on the human condition” Following this path, a UXD could go to a client and say ‘I can get your customers to do XYZ and they won’t even know we made them.’ Granted, I would say the designer in this case is being unethical, but the opportunity still arises.

    @Laura – Bottom line, it will always be the people who use the design product. If you don’t have users, guess what? You’re out of a job. That scenario could be what will always keep us from becoming true ‘puppet masters’.

    @ Adrian – Users owning the experience is an interesting concept, one I would love to explore more. Regarding context, it can be subverted though. The bow and arrow had a great utility for bringing down game, until someone thought to use it against their neighbor. So the question becomes, should we pay more attention to the outliers that could use our design badly and put some effort into preventing that scenario.

    @Brian – How much should designers look out for users? Is it ok that we stand aside and let a user burn their hand a time or two on the stove so they learn the dangers of a hot surface? Or try to prevent the situation all together? Perhaps the old standard holds true, UXD is meant to improve the human condition. I would take it a step farther though, and add UXD shouldn’t coddle human nature.

  8. Pingback: links for 2009-07-24 | burningCat

  9. Pingback: Twitted by AshishGurung

  10. I don’t get this point in your conclusion:

    It appears that we have painted ourselves into a corner. Having direct control over another person’s behavior is wrong, as they did with the watchclock.

    Wrong is a strong judgement.

    For years, dentists have been trying to convince parents to help their children brush longer. Longer teeth brushing sessions directly correlate to better oral health and fewer cavities. Yet, the dentists failed to make any progress.

    Then Dr. John’s Products released a line of children’s power toothbrushes. (Subsequently acquired by Proctor & Gamble.) The battery powered devices only have an On switch and automatically turn off after 3 minutes. The 3 minute run time forces the child to brush the entire period. Children who use the toothbrush regularly demonstrate substantial better long-term oral health than children who don’t.

    The design of the toothbrushes explicitly influences the behavior of the child, guiding them to better oral health.

    Is that wrong? shows users their spending and investing habits in a way that, for many users, changes their behavior to spend more consciously and invest more savings.

    Is that wrong?

    Sacremento’s Municipal Utility District found that when they put smiley faces on the bills that of residents who outperformed 100 of their neighbors in homes of similar size that used the same heating fuel, those households reduced energy use by 2%. The design of the bills influenced the energy use of those individuals.

    Is that wrong?

    I’m not getting how directly influencing behavior is wrong.


  11. @Jared – I agree that directly influencing behavior is ok, mostly because we can’t avoid it. Direct control is different. People who use something that has ‘control’ over them are no longer users, but pieces of the machine. The question becomes, where is the gray line that separates influence from control?

  12. Wow, I didn’t know this debate was happening! Thanks for riffing on the Watchmen piece, Brad.

    As the author of the referenced article, I want to note that I did not imply that the watchclock’s aggressive and controlling posture was in any way a bad thing.

    In fact, I will somewhat controversially question the implied thesis of your piece, Brad: That controlling other people is inherently bad, and that giving other people freedom is inherently good. There is Jared’s example where parents and medical professionals seek to enforce their will on children, of course.

    Similarly, in the watchclock example, we are talking about another perfectly socially-acceptable power relationship: a contractual agreement between an employer and an employee. There is nothing deceptive or forceful in this equation. The behavioral control implicit in the system is not an example of a system violating an individual’s rights or dignities. It is, in fact, intended to prevent the watchman from violating their side of the agreement.

    There is no reason why an employee should see fit to execute their security duties in a way that their boss chose to implement the security system. If the boss thought that allowing their night watchmen to roam freely about the premises and use their judgement about where and when to observe, they would simply make that their policy and not buy into an elaborate watchclock system. They’d probably also have to hire much more skilled and expensive guards, too. And pay a lot more for their insurance policies.

    Some employers try really hard to give their employees a greater sense of freedom, thinking that it will encourage better performance from them. They want their team to not feel like cogs in a machine. But the sad truth is that some jobs are quite simply cog jobs. And even if it were true that all humans perform better when given freedom (which is clearly not true), an employer who thought otherwise would be guilty of poor judgement, not of oppression.

    (It’s ironic that you cite Poka Yoke as an example, because Poka Yoke is in reality the one bright shiny part of an otherwise dreary system where every other movement and action of each employee is strictly controlled and measured as much as the robots that surround them. Poka Yoke is not a recognition of human dignity and creativity and freedom — rather, it is a recognition that some tasks (error detection) are, at least for now, unable to be performed by machines.)

    While your idealism regarding human dignity and freedom is admirable, I am afraid it disregards the great many contexts in life (such as employment) where these values are justifyably, er, devalued.

    To object to this system is to object to the very existence of power relationships in which one human has power over another. While such power relationships are at times unfair and even evil, I simply cannot agree that this is objectionable in all instances. And certainly not in the watchclock example.

    Thanks for digging deeper into this, Brad. Lots of stuff there to think about.

  13. There is no reason why an employee should see fit to execute their security duties in a way that their boss chose to implement the security system.

    Er, I meant “…in a way contrary to the way their boss chose…”

  14. Pingback: DESIGN :: interaction, experience » Are you sure that we are the puppet master?

  15. Pingback: the puppet masters

  16. Pingback: Are We The Puppet Masters? | UXscape