In addition to participant observation and structured interviews, there are other ways that you can forage for design insights. For example, what do you do when the behaviour that you are interested in happens over a long period of time, or is sporadic, or both? Interviews are hard for the same reason. One effective solution in places like this is for the participant to do the capturing themselves. One common class of technique for doing this is what I call diary studies. In this technique, you give people a diary that they complete at a specific time or interval, for example every evening or at every meal. In general, diary studies are used to record a specific piece of information, like “how happy you feel” or “what you ate.” Often the diary has some sort of structure to help you complete that efficiently. You can use normal old paper, text journals; you can use still or video cameras; you can have audio recording — whatever is appropriate for the task that you’re trying to capture. In selecting from among the different media that you can use for capturing — like video versus audio versus text versus photos, analog versus digital — pay attention to the context. If you’ll like to capture information about somebody’s mobile phone, maybe you’d give them a piece of paper with some structure on it, like a couple of scales that they’d keep with them on their phone, so, that way, whenever they pull out their phone, the diary entry that you’d like them to fill out is right there with it too. In some cases, an audio recorder will be the easiest way to get people to actually record information at the relevant time, like maybe if somebody’s driving. In other cases, like during a lecture, speaking out loud might be inappropriate, and so you’ll want to have somebody have marks on paper. One of the appealing features of diary studies is that they can scale a lot better than direct observation. Direct observation is limited by the amount of time that you can spend with the participants; with a diary study, you’re only limited by the amount of materials that you can give out or that you can aggregate later on. The most important piece of design for creating an effective diary study is to have the entry be as frictionless as possible. The easier it is for participants to mark down the information that you’re interested in, the higher quality the results that you’re going to get. With a diary study, like any user interface, the results that you’ll get will be best if you offered people some training and some practice. Also, any time you’re changing people’s behaviour, like asking them to record into a diary, they’ll do it for a while, and then it’s easy to fall off the wagon and forget, and so you may want to follow up with people and remind them, and that reminding brings us to our next technique, which is called “experience sampling.” The idea behind experience sampling is to “beep” people at some regular interval, and have them write down a key piece of information at that time. Sometimes these are also called “pager studies”, because many of the early studies in the 80’s and 90’s used pagers, and an appeal of doing a pager study is that the participants don’t need to remember because you’re actively reminding them. They’re often coupled with some kind of diary, so the paper beeps — or now it might be your mobile phone — and then there’s a structured form that you’d fill in. And these are, again, used for things like “How happy do you feel?”, “What’s your energy level?”, “Where are you?” Sometimes these take the form of a psychometric scale; other times these are more open-ended questions. On the research front, technologies like wireless sensor networks are expanding the possibilities for what the triggers can be and for what kinds of information can be automatically or semi-automatically recorded. Experience sampling, like diary studies, is useful for aggregating information across lots of people, like, “Are there times of day that make people more or less happy?” It’s easiest for you, as the researcher, if this information can be filled in in some digital form, like a survey, so it can be automatically aggregated. But sometimes, for practical reasons, paper will be the most ubiquitous tool out there, and in that case, go with whatever you can to get people to actually fill it out. In the techniques that we’ve talked about so far, it’s the designer that ultimately comes up with the design ideas, and the user’s behaviour serves as the father for that ideation. Users can also be a great source of design ideas themselves, especially advanced users or “lead users”. And Eric von Hippel at MIT has been the champion of this approach for several decades. He’s in the left of this picture, and he’s hanging out with Dr. Nathaniel Sims at Mass. General Hospital in Boston. Dr. Sims is an anesthesiologist. Like almost anybody in almost any work environment, he finds some of the tools that he has to use frustrating. But he went one step further than most: For example, when he needed to carry around a number of different medical devices, he created for himself a carrying rack that could easily hold all of them at once, so they can be moved around the hospital more efficiently. He’s picked up several patents for his work over his career, including this device here, which is called the “Nat rack”. Lead users in all sorts of domains come up with clever solutions, and one role of designers is to help lead users turn their individual solutions into something that’s more generally useful. And, in this way, lead users become a sort of distributed creation engine who can collaborate with designers to bring products to market. Lead user innovation works best when the reason that there’s not a better solution out there is primarily because designers don’t understand what the user needs are, or the context is shifting really rapidly. And so, for example, Eric von Hippel has shown how in places like surfng and snowboarding or other extreme sports that move quickly, changing your equipment is not all that difficult, but things are fast-paced, and so, to be able to do new tricks, people will modify their equipment to suit their needs. Lead user innovation works less well when the necessary piece of information is some kind of process knowledge, or a better factory, or something like that. Related to lead users are what we might call “extreme users”. Think about something like email. All of us get a lot of email, but some of us get a lot more than others. Those people who get a whole lot of email, far more than the average person, they’re extreme users from the vantage point of email. And we can often learn things from those extreme users — how they handle thousands of messages a day, for example — that we might then be able to encapsulate and make available to all users and help everyone. Extreme users can be extreme in almost any direction, and so people who have interesting professions are often a good source for extreme users. One can be an extreme as a technophile, or one might be an extreme as a technophobe. And so the person in the log cabin in Vermont, who checks email once a month, might be as useful an extreme user as the CEO in Silicon Valley who gets thousands of messages a day. While lead users and extreme users can often provide valuable design ideas that transfer more broadly, it’s not automatically the case, and, in fact, sometimes the extreme users are extreme because they’re not the actual users. Make sure you keep in mind the actual people that you’re designing for. You do all this great design work at the beginning to learn what users need. How do you keep their needs in mind throughout the entire design process? How do you not lose track of these insights that you captured early on? One great strategy for distilling the insights from participant observation, or interviews, diary studies, experience sampling — any of the techniques you choose — is to create from those insights “personas”, who are abstract users who represent what you’ve found when you went out and looked at real users. So, a persona is a model of a person; they’re an example. They’re not any one human being, but they are concrete. So, a persona is going to include demographic information, and also their motivation — Why do they want to use the system that you’re creating? What would make them not use it? What are their beliefs? What are their intentions? What are their behaviours? and what are their goals? What often happens in a design process is that one design member of the team want to build something, and so they’ll make up a story about why that particular thing might be useful to somebody. A persona keeps you grounded. You can say, “How would Steve use this?” or, “Would this additional feature fit with Steve’s desire for a minimalist system?” To make these personas real, it’s nice to have a picture or a photo. In fact you can use stock photography, or one of the photographs from your needfinding to anchor that persona visually. Make sure to give your persona a name, give them an occupation, and a background. They should have some hopes and dreams. Give them a story to tell. They should really come alive and feel like a real human being. It’s easier to be empathic towards a particular person than a generic one, and that’s how personas help: By knowing what a persona thinks, does, and feels, it helps you build empathy; it helps you understand the states of mind, the emotions, the philosophy, the beliefs, the point of view of that user. Personas also keep designs coherent and consistent over time, rather than a scattered-shot agglomeration of features. And, perhaps most importantly, the empathy that you’d build by designing for a particular persona can often lead to insights that you wouldn’t otherwise have, and this gives you new design opportunities and can help you be more innovative than existing solutions that are out there. We’ve talked about several strategies for engaging the people to come out with new design ideas. This is the best way that I know how to reliably come up with innovative ideas. But that doesn’t mean that every single design has to work this way — every design process has to work this way — and it doesn’t mean that, automatically, if you failed to follow this design process, then your design is automatically bad. It’s not like, if your startup goes public, the SEC, on its filing forms will ask you “Did you follow a rigorous needfinding process?” Ultimately, what people are excited about is the design, and all I’m offering here is a set of tools that will help you, with the best odds that I know of, give you as as great a design as you can get.