In part 3, I talked with Dr. BJ Fogg about conducting research on persuasion and incorporating the results of that research into projects. Be sure to read part 1 and part 2 for more on persuasion and designing emotion…
TvG: I’ve talked with some designers who’ve used persuasive methods in design projects and missed the mark because they’ve focused on the wrong level of the functional triad. How we can go about identifying what the most crucial elements or levels are for persuasion?
BJF: That’s a good question. Step one is to get really clear about what your persuasive goal is. What attitude do you want to change or reinforce? What behavior do you want to change or reinforce? You can’t do everything. There’s a process that I do with clients to help them decide on their persuasive goals called Impact Analysis. It’s a systematic way of saying ”what do you want to do with this website”? You can’t do everything. You can’t achieve forty goals. You can achieve two or three. If you focus it on one, that’s even better.
You have to really crystallize the team around that goal. In almost every case, I’ve worked with companies where they haven’t decided what the goal is yet. They’ll say something like “our VP says we need to be doing something in the social networking space.” What I do is to get the team very clear in term of what their overall persuasive goal is.
Step two is when the creative work begins and you can brainstorm, or go look at examples where other people have achieved that goal. There are lots of rich examples in the web 2.0 world where companies have successfully leveraged social software to achieve persuasive outcomes. The safest approach is to copy what has worked before.
Then, you prototype and test it. A lot of the time when people are new to designing for persuasion they’ll think “okay, we’ll do this, and then this, and then this, and people will perform this behavior”. But the most successful persuasive systems don’t even feel like they’re persuading you. Persuasion needs to be designed with the lightest touch possible. What you want is the most minimal intervention necessary to get users to perform the target behavior.
If you start layering in strategy upon strategy, incentive on incentive, it feels heavy and artificial and tends to turn people off. A good example of a lightweight system is Ebay’s feedback system; little stars. It’s very easy and not framed as a persuasive system, but at the end of the day that’s what it is. It persuades you to trust a stranger enough to do a financial transaction. They call it “feedback”, but that’s what’s made Ebay an empire; those little stars are the persuasive system.
In terms of placement on the functional triad, I advocate brainstorming in all three corners; Tool, Medium and Social Actor. Now that there are many examples of things that work, it’s often a case of taking a look at what’s worked before with the audience and then tailoring that solution to your situation.
TvG: Does your research on persuasion encompass any techniques that you would say are outside of normal user experience research methods?
BJF: I have some methods that I wouldn’t call research, like my Impact Analysis. Last I counted, I’ve created over forty frameworks that take fuzzy psychological dynamics and make them concrete and actionable, such as how to achieve delight or simplicity in product design. I usually create these frameworks by synthesizing scientific theories and my insights about user experience with technology. Generally, my frameworks help design teams to see what matters most.
Just as important, the frameworks give people a similar point of view and specific language for discussing these fuzzy psychological dynamics. For example, if the team is trying to create a website that conveys “trust,” one of my frameworks defines this vague word and breaks the concept into understandable parts. Then when people talk about “trust” they are talking about the same psychological construct. That helps a lot.
TvG: Can you tell my readers more about one of your frameworks?
BJF: After doing three years of research on web credibility, I synthesized my lab’s findings in a framework called Prominence-Interpretation Theory. This theory says, “figure out what elements people interpret as credible on a web page and then make those elements prominent in the execution”.
In practice this means finding out what makes your page credible. Maybe it’s an association with a university, or showing you have data to back up claims, or that you have 5 million users. Then take those credible elements and make them more noticeable.
Prominence-Interpretation Theory is simple, really. It also guides the design process by showing that you need to answer some key questions when you design for credibilty, such as:
1. Who’s our audience?
2. What do they associate with credibility?
3. How do we make that prominent in the experience?
You can test each step along the way. The process is clear and linear.
TvG: So if I were to describe up Prominence Interpretation Theory, I would say that you’re making certain elements more prominent in order to draw increased attention. How much of persuasion would you say is about directing attention from one element to another?
BJF: Good question! I think what you’re really addressing there is framing. So how much is persuasion about framing? Framing tells people how to interpret the situation. It’s that subjective interpretation that makes persuasion easy or difficult. A huge part of persuasion is about directing attention by framing and aligning people’s thoughts so that they see the picture the way you want them to.
TvG: I’m also thinking of terms of things like hierarchy in graphic and information design. Variations in size, colour and contrast are ways to indicate prominence. In that way, I would think that Prominence Interpretation Theory could be applied to the whole world of design.
BJF: It even comes down to how your write the sentences. “We’re going to start charging for services”. There are ways to frame that statement more positively. “We’ve reached a point where we’re adding new features and we’re now charging this low fee.” Part of it is visual, part of it is the hierarchies, the outlines. Part of it is the rhetorical stance of the text and the tone. It all has to come together to achieve that persuasive goal. Going back to my Impact Analysis process, if you know what your goal is, on the visual level, on the structural level and in the writing, they all need to converge to achieve consistency with one or two persuasive objectives.
TvG: Your colleague Clifford Nass showed how people attribute personalities to media in The Media Equation. Lately, I’ve been thinking and writing about product and brand personalities. The product personality needs to be consistent throughout the user experience, including product itself, the interface, the packaging and advertising. This needs to occur even though the average product is the result of the work of hundreds of people, in order to convey a consistent personality.
BJF: One thing I like doing with industry clients is quick and dirty surveys where we put things like logos and layouts on the internet and get quick feedback. Whether it’s emailing twenty friends to find out which logo they like better, or spamming a few hundred people, we can always get some good information.
But, as I said, what I love doing most is creating a new psychological framework. Taking something fuzzy, and organizing it. It gives everyone a way to think about the topic and a common vocabulary to discuss the project. Over the last couple years, I’ve been developing a framework for simplicity. I’d say the beta version is done. My framework breaks simplicity down into discreet dimensions and quantifiable parts.
Like I said before, when people work together on a project, it’s important that everyone is talking about the same thing. In user experience, words are a big problem. I talked about “trust” earlier. This word can mean a bunch of different things, from technical robustness to personal credibility. Because “trust” is such a common word, most people don’t realize that it’s a complex psychological construct. So shared vocabulary is very important. For me, it goes back to taking fuzzy things and making them concrete.
TvG: How do you discover what a particular audience group finds credible?
BJF: In the real world, in most cases, companies are guessing. Few companies have the patience to go out and sit and talk to people. Companies always think “focus groups – eight to twelve people in a room”. I’m not a big fan of focus groups. I’ve done that for clients but I usually say “I’d rather bring in friendship pairs for discussion interviews and here’s why…”. I’m a big advocate of doing friendship interviews. Two people who know each other come in and you talk to them.
Sometimes it’s just a matter of helping clients to think clearly about the problem. Depending on how well you know the client, you have to ask questions. “Who’s your audience? What part of that audience is really going to the website? What do they find credible?”
In the case of a medical site, it might be providing references beside the information. In the professional world, you have to find ways to do quick and dirty discovery that answers those questions.
TvG: So what is the response from industry to this kind of research?
BJF: It depends on the client. Startups need to move really fast and they’re usually guessing. The good news is that great startups can change their directly quickly if they’ve guessed wrong. For a larger company that has a two-year horizon on product development, they have the luxury of doing deeper research. But big companies don’t often move quickly – or change their direction if something’s not working as expected.
If a big company is already planning focus groups. I won’t change their mind about that since it’s already budgeted. Sometimes I’ll attach my piece to those focus groups. You have to understand that companies have personalities and ways of doing things? and you aren’t likely going to be able to change company culture. So you have to pick your battles.
TvG: Who are some of your heroes, people who have inspired you over the years?
BJF: Terry Winograd is someone I admire. Terry is very smart, very forthright and well-intentioned. Cliff Nass is brilliant; he continually impresses me. Phil Zimbardo is remarkable for the way he brings psychology to the masses. I admire the way Malcolm Gladwell and Donald Norman write. Generally, I admire people who have positive impact in the world.
TvG: What are you currently reading?
BJF: I’m reading Web Analytics: An Hour a Day and Getting Real. I’m re-reading The Tipping Point and I’m listening to Rich Dad Poor Dad on my iPod. I’m re-reading some Steinbeck, just for fun. I’m also reading material by John Marrow right now. I was invited to be on the board for Learning Matters, and John is the founder of that organization, so I’m ramping up.
Let’s see… I’m also reading Persuasive Games by Ian Bogost. Frankly, I have many books waiting for me. Thanks to Karen Pryor, who I invited to speak at the recent Persuasive Tech conference, I have two new books. And she’s pointed me to books like The Emotional Brain. I have some persuasion books I need to be reading but haven’t yet. I never seem to get enough time to read. Last I counted, I have 63 books waiting to be read; I’ve stacked them in eight piles clustered by topic and how carefully I need to read them.
TvG: What advice do you have for designers looking to incorporate persuasive emotional design principles into their own design work?
BJF: #1. Learn empathy. I think that’s the most important quality for a designer. Some people have more empathy, some have less. I think it can be developed to some degree, but if you’re not an empathetic person, you probably won’t be a great designer or user experience professional.
#2. I think it’s very valuable to understand the deep, hard wiring of the human brain so you get a sense of what drives human behavior. That’s why I’m reading books like Don’t Shoot the Dog, The Emotional Brain and The Moral Animal. Get a sense of what drives us as creatures. Sometimes you need to go through a lot of complexity, but try to distill it down to a fundamental understanding of what drives human behavior. A lot of it is social dynamics. But it’s really not that complicated. I think having a gut sense helps.
#3. Develop a range of research methods, not just one thing. And for industry, they have to be quick and dirty. As in overnight!
TvG: Well, it’s been great talking with you BJ! Thanks very much for taking the time to speak with me today!
BJF: No problem Trevor! Fun to talk with you!
For more on persuasion and emotion, see:
An Interview with Dr. BJ Fogg – Pt. 1
An Interview with Dr. BJ Fogg – Pt. 2