Fun Theory and Behavior Design

Behavior Design field marries design and behavioral science and economics. Today, Fun Theory links up with game design, since it focuses on understanding, quantifying, gauging the impact of funs (as the opposite of boredom). It tries to answer problems such as how we should quantify of fun. Of course, it has strong links with behavior design since it maintains that routine behaviors can be impacted by introducing elements of fun. Volkswagen made the concept of Fun Theory when it promoted experiments that proved the theory. The experiments also constitute great examples of behavior design. In fact, fun theory can be seen as a strategy in designing behaviors. 

One important aspect of behavior design is the Fogg Behavior Model (FBM) The FBM has three-parts. For a behavior to occur three elements must be at play: Motivation, Ability and Trigger. Motivation is self- explanatory: what do I expect to gain or avoid by engaging in the behavior? Ability are the aspects of the current context or design that enable –or not– a behavior. Lastly, triggers are reminders, something in the environment that indicates that a behavior should happen now. The aspect of kairos, greek for opportune persuasion, is what needs to be remembered when looking at trigger. Is the trigger coming at a time when motivation, ability and trigger are working together to ensure the behavior?

Let's look at three Fun Theory experiments through the FBM.

  1. Bottle Bank Arcade Machine: This idea makes the return of plastic bottles and cans a fun activity. In this context there is no payment for the return of glass, as there is with things like cans and plastic. The video shows an arcade-type container that makes sounds and provides points to the gamer. The result is the container gets used 500% more times. By incentivizing the recycling of cans and plastic yet not doing the same when it comes to glass the designers of this behavior (planners) are possibly making the activity pale in comparison. The behavior of recycling plastic and cans become something that people remember to do because it's an easy way to make pocket money. And some people might be thinking that it's unfair to get not money for a very similar activity for which they do get money. The experiment in the video shows an increased result during one night of usage. The problem is how do you make that game sticky so there will be a reason to use it more than once. Without knowing the game itself and what long term potential it has, it's difficult to ascertain what long term usage it will have in the community where the machine would stay.  Looking at the  FBM, there's a lot right with the design in terms of the trigger. The container is noisy and flashy (literally) and in the short experiment, they are shown to attract attention. However, these to aspects could become annoying with time because they are triggering (reminding) everyone independently of whether the passerby can possibly address the trigger. He might or might not have glass bottles. So the kairos might be off. 
  2. The Worlds deepest bin. A bin is rigged so that when used it makes a sound that makes the person outside think that the object is falling through an extremely deep hole.  As a result, it collected 41 more kilos of trash than a nearby bin.  This idea is working, like the previous one, on the curiosity of passersby. If this area has a large number of people who are first time users, for example if the bin is located at a tourist attraction the design is likely to increase collection long term. A better solution could be letting people know that other people threw the trash there. Knowing that others engage in a behavior makes it socially sanctioned and is one of the most potent strategies for behavior design. 
  3. The play Belt: This idea restraints passengers in their ability to use the entertainment centers in the car. They can only do so when they use their safety belts. My thoughts: strictly speaking, this is what behavioral economist call paternal libertarianism, which is the belief in designing "nudges" into products and services that push the user to make the right choice. I would posit that this behavior is not fun per se, but that the prize for engaging in the behavior is. 

In conclusion, though these examples show that you can change behavior through fun, there will frequently be better ways to attain the same results by thinking beyond fun an at a behavior long term. I would say that in these three examples, I have no reason to believe that there was actual long term behavior change (learning) taking place.

The Reiss Motivation Profile as a behavior design tool

The Reiss Motivation Profile was of particular interest to me as someone whose area of focus is the intersection between services and behavior. During the writing of my thesis on the subject I used a model of behavior, the Fogg Behavior Model (FBM), that covers motivation and further divides it into three kinds: hope/fear, pain/pleasure and social acceptance/rejection. I think the RMP is a great attempt to further personalize the aspect of behavior and use it, as stated on the paper for anything from sports coaching to marriage counseling. 

The positive aspect of the RMP is that there is personalized detail about the individual. The major failing point that I see is that (according to the FBM) behavior does not depend only on motivation, but on ability factors and on triggers, the reminders of appropriate behavior. Therefore, used by itself, the RMP might fail as a tool to generate a target behavior.

Here is an example of this incompleteness as a behavior design tool, taking from my own RMP results. One of my highest motivations, my third, is the desire for beauty. And indeed, I will take a lot of steps to insure that my environment is beautiful. But, together with having high need for beauty I also have limited funds. So my ugly, second hand sofa, that I bought from my friend  years ago is still with me. And it will still be with me until I feel that I can buy a new one (using cash) that is much more beautiful. The lack of funds, of course, is one of the 6 ability factors and it's, right now, trumping the motivation. 

That being said, this tool, used together with the FBM, could be very useful when designing a behavior for a specific person. Services that are personalized and both seek to change behaviors will become increasingly popular (gyms, healthcare, banking).  Deepening the understanding of an individual's motivations before using it to create services that allow the him or her to change behavior will eventually be common-place.

It could also work, partially, by assigning one of the 16 motivations in the RMP to a specific target group. For example, if one is designing a gym for women in their 20s and 30s, one might use the RMP as a research tool and provide to a number of research participants. If, for example, we find that a great number of them are motivated by romance, family and social contact, this will undoubtedly need to impact the design of all aspects of the gym, from communications to business models (do we tie the gym to a dating service?) to what specifically the gym does to modify a behavior, for the most part, using the gym frequently enough to produce results.

Screen Shot 2017-12-17 at 4.09.15 PM.png

It's the behavior, silly: Using Behavior Design to evaluate technologies

It’s 2003. Right before the commercial break, the camera cuts to Ryan Seacrest, who is sporting radiant frosted tips and is sitting by a girl, one of the final 10. She frantically waves to the camera and grins like a maniac. On Seacrest’s hand a cell phone. He energetically explains to the audience how to vote for their favorite using texting. The camera shows a close-up on the phone’s greenish screen as he clicks three times on the “8” button to get to “v”, for “VOTE”.


In 2003 American Idol collaborated with AT&T (then Cingulair) to create a voting system that used text messages. At the time texting had not been adopted except by the younger crowd. The “cut-to-Seacrest” routine described above repeated itself many times. Seacrest taught America to text. This is not an overstatement. A 2008 poll by AT&T shows 22% of respondents cited wanting to vote on American Idol as the reason they learned to text.

How was American Idol able to have such impact? What can we learn from this case study that is applicable to other technologies? This becomes clear if we see this story through the Fogg Behavior Model(FBM). The FBM is a three-part model. For a behavior to occur three elements must be at play: Motivation, Ability and Trigger. Motivation is self- explanatory: what do I expect to gain or avoid by engaging in the behavior? Ability are the aspects of the current context or design that enable –or not– a behavior. Lastly, triggers are reminders, something in the environment that indicates that a behavior should happen now.

The American Idol “vote-by-texting” behavior example can be charted on top of the three elements of the FBM. First, think about the fervor many viewers felt for their favorites. Teenagers, grandmas and tough-looking males would organize viewing parties, get t-shirts and scream like they were possessed if Justin or Kelly advanced to the next round. This is clearly a high-motivation situation: they really (really) want their favorite to win and know that they can influence the result by texting. Next, the six ability factors that the FBM presents are all either intrinsically simple or are made simple by the show. This is a behavior that does not take a long time or cost a lot of money or require great physical effort and one that is socially accepted. The two factors working against it are that –at the moment –the behavior is non-routine and a there is a learning curve to consider. Enter Seacrest. He’ll explain it to the audience enough times so texting becomes a routine task. The last element, triggers or reminder– is more than covered by the show being on multiple times a week for many weeks.

The title of this post presents my biggest aha moment since I graduated from SCAD as a Service Designer and did my thesis on Behavior Design. Which, admittedly, is not too long ago. What surprises me that is that as much as I lived and breathed the idea of Behavior Design during the writing of my thesis, it was thinking about the adoption of new technologies within services that made the FBM model come vibrantly alive.

A more current example of designing for a behavior that involves technology are behaviors that require the adoption of VR. Some experts have started to doubt whether customers will ever care at all about VR because -contrary to its next-big-thing promise– it hasn’t been immediately and widely adopted. They focus on the technology. However, if we run the behavior through the FBM we can see large problem areas on the behavioral side. Let's focus on the social aspect of motivation. The FBM drills down into three kinds of motivation. One of them is a combination of social rejection and social acceptance. People will be more likely to adopt a behavior that avoids the first and increases the later. It is clear that there’s an anti-social aspect of sporting a mask in public. Also, depending on the virtual world the user is stepping into, he might be alone in there too.

Does the finding of hurdles for its adoption mean that VR is a tragically flawed technology and should be put aside? Should we decide that it can only be used by gamers and never by, say, users at a museum, an airport or a supermarket? Not really. Behavior Design can step in and test iterations that intend to make VR more social.

Here’s a funny meme on QR codes, titled “Helpful flow chart”. At the top the flowchart starts: “Should we use a QR code?” The arrow points only to “No”. The next arrow points to a “But what if..?” Yet another arrow bluntly closes the subject. “OMG, no!!” I’ve joined in the snickering about QR Codes. I’ve rolled my eyes and enjoyed being a skeptic. But the fact remains that QR codes have been widely used in much of Asia. And that QR Codes are happily used thousand of times every day in American airports. Could it be that QR codes are a perfectly viable technology that was misused and abysmally implemented by designers in the US? It’s likely. This is one of the thoughts that halted my snickering and changed my attitude.

The attitude designers should adopt is a measured agnosticism about technology and, at the same time, a belief in their own ability to design for behavior involving a new technology. This is not to say that any technology will work. Or that the technology won't need to be adapted. What I'm saying is that it's necessary to be skeptical about it and hold it up against the elements of the behavior needed to use it.

Let’s go back to our texting example. In 2003 a smart behavior designer would have recognized that the public didn’t know of a strong reason (motivation) to abandon phone calls, which it knew how to do, in favor of texting, which it didn’t know how to do. Until, of course, it became the only way you could vote for your favorite American Idol. Our designer would have looked at the ability factors. Is texting really that difficult? No. Does it work as advertised virtually every time? Yes. Is it less time consuming than calling someone up? Potentially. Therefore at that time it would have been smart to assume that texting was neither a failed nor a perfect technology, but that it needed to be designed into a behavior and that an appropriate level of effort had to be invested into motivating, enabling and triggering consumers. Which is what the show, together with AT&T, did.

Behavior Designers and Service Designers need to be agnostic when evaluating using a technology to promote a desired behavior. Three questions are crucial: First, does this technology benefit the user? Second, does it generally perform as promised (it doesn’t fail, doesn’t have a major bug)? And third, is the enterprise willing and able to support the effort behind creating the context in which the behavior is likely to happen? A behavior that uses a technology can require such an investment in communicating how to use it, for example, that implementing it without said investment makes no sense. In that case a designer is justified in discarding it for that behavior.

In a world of such breath-taking technological advances is easy to be dazzled and join the crowd ready to pass judgement and declare winners and losers. My biggest lesson, as of now, is to always remember: “it’s the behavior, silly.”

A tried (once) and true (at least for me) thesis-writing workflow

I am done. All defended, sealed, delivered. Having accrued some experience in how to make all possible mistakes in writing a thesis, I thought I should write a post on the workflow that I eventually came up with. I wasted 20% or more of the total time invested. Here's my tried (once) and true (at least for me) ideal workflow for writing a thesis. 

1. Have a clear outline or list of core ideas. For me, thinking of ideas was more useful than doing an outline because I didn't have to commit to the final order of the outline until later. Examples of core ideas are: "Humans have limited willpower" and "The Service Design Community of Practice has started to write about behavior design".

2. Whenever you read always, immediately type your notes on a computer and save days and days of work. If you do not do it right away, you will have to come back to it. I understand you might do some exploratory reading in the beginning, but after the moment you set up your core ideas always type your notes up. 

3. As you are reading and typing, keep asking yourself: is this quote answering one of my core ideas? If not, it is useless to you. Leave it alone. This is why you need to start by knowing your core ideas. Because otherwise you will write down things that are interesting, fascinating, beautifully put but of no relevance for your thesis.

4. Now the question is: what to type? You should always write a. The exact quote (except if you know, I guess because you got a signal from a higher power, that you will not use it as a quote but a paraphrase) b. The paraphrasing of the quote. If you can do it as you are making the note, you are saving untold amounts of time. I was not always able. c. The page in the book. I lost so much time on this, just in the small number of times I did not register the precise page in the book. d. The APA style (or whatever style you agree with your thesis chair) in-text citation. So for example, “Quote from Kahneman, bla, bla, bla" (Kahneman, 2011, p.302)“. This way when you are putting everything together, you already did this. I recommend learning how to do citations on the style you are using before ever starting taking notes of anything. Which alas, is not what I did.

5.  Digital or paper? Many prefer to touch and *be one with the text*, but this might be the moment in which e-books make more sense. In some cases, I even doubled-up by checking out the book from the library and buying an e-book. As your read above, I am proposing you have the complete note, including an exact quote on your computer as you are reading. I will argue that having an e-book on your computer will let you copy-paste the quote into whatever word processor app you end up using.

6.  Now, on how to organize the notes. Try Gingko, ( It is a horizontal outlining app. Gingko grows horizontally, and you use cards that you can move around. The way you use it is you put all of your core ideas in cards (you can move them around later). Then, write your complete note (see tip 4) as a child of the particular core idea. The beauty of using this method is that when you finish you export the Gingko board, and there it is! You will have your literature review almost done and in the correct order. The only thing left will be writing text to link all the notes with introductory texts. This method works for all chapters, of course, but is most useful for a literature review. 

7. Organize your sources. You will scan the literature. You will skim through many books. You will not necessarily use all of them (see tip 3). If you do not organize your sources, you will go crazy. Below, a link to my database of sources.  I always go for more rather than less, but of course, you can adjust this database ( to your particular level of type-a-ness. 

8. Learn how to use the paragraph styles in Word and set up your paragraph styles so that later you can make changes to the styles and not to the text. Paragraph styles are the styles of the body copy and the different levels of heading. The knowledge will save you time and is important when doing your lists of figures and your table of contents.

9. Use to check for good grammar and plagiarism. It also has a vocabulary enhancement feature that keeps you from repeating words and suggest substitutions. It is a paid service but it's worth is for the short time you will be writing.

That is all. Good luck!

My top Behavior Design/Change books

After scanning the literature for books on the subject and skimming around a 100 sources and reading around 50 of them, here's my top 13 list. This is bound to be a fluid list, since I continue to snowball from book to its sources, but for now, here it is!

Goals for my knowledge tool for the Fogg Behavior Model

I am doing a knowledge tool for my thesis. It's not the final artifact, but the method I will use to co-create. The thesis (of my thesis) is that behavior is the most important currency for the design of services and that we need a shared model and a way to use that model to systematically understand behavior.  

Here are the specific goals for the knowledge tool (K-tool)

  • Establish a shared definition of Behavior for the purposes of the K-tool
  • Communicate why understanding behavior is so important for services
  • Explain the three drivers of behavior in a general fashion
  • Go deeper into each driver
  • Review, test and get feedback on the transfer of the information
  • Present a Service Design Tool based on the Fogg Behavior Model
  • Present a Case Study within which to apply this tool 
  • Get feedback on the tool


The Fogg Behavior Model

At the center of my thesis is the idea that Service Design (and I add here, all design thinking fields) lacks a clear, shared, behavior model and that this fact is detrimental to our work. As part of my research I will be conducting workshops with  students and practitioners to co-create what could come out of incorporating the Fogg Behavior model. 

As part of these co-creation efforts I am creating informational tools that will help me communicate the model. One of these is a presentation of the model. I share it here.

Behavior mapping for my Information Architecture Class

In a great double-dipping maneuver, I made my Information Architecture personal data project match my thesis subject. The directive was to come up with seven (not necessarily related) variables, collect data for eight weeks and then visualize the resulting dataset. I extended this main objective for this project to help me learn how to map behaviors. More specifically I want to map just one behavior: whether I exercise outside every morning after waking up. 

For my thesis I am working with a three-part behavior model that posits that all behavior requires three elements to be present:

  • Motivation
  • Ability
  • Trigger

The model further expands each of these elements. According to BJ Fogg, the author of the model, motivation can be divided into three core types:

  • Pleasure Seeking / Pain avoidance
  • Social acceptance/ Social deviance
  • Hope/ Fear

There are six different types of ability factors. These are conditions that enable an action. They include time and money, for example.

Lastly there are three types of triggers, or reminders.

Of course, after seeing this model, you probably suspect it is not about just logging whether I did or not in fact go out each day. That would be as easy as logging date and yes or no for the exercise. It is about figuring out what factors correlate to my ability. It’s about measuring motivation. It’s about setting and following triggers, or not.

I will post more on this project, as it is closely related to my thesis.

For the class we had to present a hand-made visual for our idea. Here's mine.