I've talked a lot about user needs, and I've shown you some nice lists of user needs. I want to turn to the question now of exactly how we come up with those nice lists. And the method that I'm going to describe is a little bit detailed, so I'm going to switch from this video format to annotated PowerPoint. You've heard me say before that problem definition is usefully divided into two pieces. First, an articulation of the overall problem statement, and then an identification of user needs. I've talked in another video about how to use the five Ys method to articulate the problem statement, so here I'm going to focus on the actual mechanics of how it is that we identify the user needs. The, the needs, themselves, look like this. This is a list of 50 of the 66 needs for the Urban Cart. The list is, is presented in all of its detail, all 66 needs in Chapter three of the design book. And for space considerations, I can only show 55 here. If you think about it, these 66 needs are a reflection of a lot of data gathering in which the designer observes and interviews users in order to figure out what are the qualities or characteristics of the artifact that would close the gap in the user experience. Now, where do these come from? And how do you identify them? And how do you get them in this form? There are five basic steps to doing this. The first is to connect with the users. And I'm going to talk about two ways to do that, observational methods and interviews. The second step is to collect the raw data based on that connection with the user. The third step is to code that data into statements of need or need statements. And, there are some methods and tools for doing that, some conventions for doing that. Fourth is to flag those needs, which are the latent needs, Which I talked about, when I talked about the Kano Method previously. And lastly, in some settings, it's worthwhile to go ahead and, and establish the relative importance of the needs, either subjectively or with some survey methods. So, let me take these in turn, and in this first video I'll get to the, through collection of raw data, and then, I'll pick up on the coding in the subsequent video. The first question in connecting with the users is who do you connect with and how many users do you connect with? In general, you can think about users as falling into distinct segments or categories. So, for instance, for the ice cream scoop, you might imagine home users and you might imagine professional users. Those, We typically refer to as segments or user groups and you need to really think of them as being distinct because the needs, for instance for the consumer segment, will be quite different from the needs for the professional segment. Within each segment, you need to connect with a number of users. Any single user will have needs that are idiosyncratic to him or her and your goal is to find characterization of the needs across a pool of users or a collection of users that share common needs. And as a result, you'll really need to talk to a number of users. These plots here, were created by Abbie Griffin and John Hauser, and a paper called The Voice of the Customer. What they did was they surveyed a very large number of users, let's say, a 100, around a particular product category. I think it was picnic rulers. And then, they looked at how many of the needs that they eventually identified were identified as a function of the number of people that they talked to. And they used two different methods. They used one on one interviews, which are shown here as the blue dots, and focus groups, which are shown as the red squares. A focus group is simply a collection of users that, that participate as a group in a discussion. Usually, it's six to ten users that sit around a table and participate in a group discussion. The interviews were conducted over one hour and the focus groups were conducted over two hours. So, Each focus group represented here had twice as much interaction time as the interviews. And the first thing to observe from these data is that focus groups are actually no more effective per hour than one-on-one interviews. In fact, they're less effective. So, for instance, if you look at spending just the first hour, One hour, Then you would identify, say, about 35 needs in one on one interviews. And you identify, identify about 55 needs in twice as much time, That is two hours, and with a group of six to ten participants. So, per person hour of interaction, a focus group is actually very ineffective as a way to identify user needs. One-on-one interviews are, are almost always better and are the most efficient way to identify those needs. So, you're going to want to interact with users one-on-one And if you look here at the number of needs that are identified as a function of the number of respondents or groups, what you see is that by the time you've interviewed seven, eight, nine users, you're up around 80 or 90% of the needs that will eventually be identified. So, as a rule of thumb, if you're in a setting where the project is quite important, getting the right answer is quite important, you want to interview a minimum of ten users in each distinct segment, and that's going to give you about 95% of the needs that you would eventually identify, say, in interviewing hundreds of people. So, as a rule of thumb, Plan on, plan on identifying or discussing the needs with about ten distinct users in each distinct segment, and in, as a, in, in more informal efforts where, say, the stakes are, are less high, you can get away with as few as about five, this five users in each distinct segment, and you'll have identified about 80% of the needs that can eventually be identified. So, there are two basic ways to connect with users. First, are observational methods and basically an observational method is a passive observation of the users in action. As a general rule, you want to go to the users use environment. That is, the environment or location in which their actually using the product or the artifact. And the things you're looking for are cues as to what the needs are. What are the gaps in the user experience? So, you want to look for things like awkward actions. You want to look for situations in which the users have actually gone and. And made solutions with tape and glue and homemade objects to. Solve their problems. Yeah, signs are actually and I literally mean, you know, handwritten signs on paper or cardboard are really good queue that. That there's something wrong. That is, that something isn't working right when a user has to make a sign to alert others as to some problem with the system in which he or she operates. You can often see visible frustration, inefficiencies or errors propping up when you just observe in the use environment. Now, I want to show you here a little bit of video, that I shot this afternoon. There's a, A really nice ice cream store near where I live in Philadelphia called Franklin Fountain, and I just went in to observe some ice cream scooping in use. This is an example of observing the scoopers in their use environment. They were actually a little bit unfriendly, honestly. And there are also permission issues in showing their, showing the, the actual people here on a public video. And so, I just, I just shot a little video of the outside. What you would do is you would go in and, and buy some ice cream and you watch very carefully what they were doing. You strike up a conversation with them, and you discuss with them their frustrations with the existing methods. Now, I, I obviously didn't show you very much of that, but you could spend an hour or more in that shop, and if it isn't a, weren't a busy morning, you'd be able to have probably a quite rich discussion. And I, I don't mean to be disrespectful of Franklin Fountain. I can understand why they don't want people videotaping in there. But so I understand why they made the decision they did today. But that just gives you a feel for the way observational methods work. Now, of course, observational methods are much easier when you're designing an artifact to address an existing problem, for which there are existing solutions. I could go in and talk to them about their existing ice cream scoops, and they're already scooping ice cream. And so, that's actually a much easier challenge than when you're trying to design something that's never been created before, or around which your users have no familiarity or limited familiarity. As a result of that interaction, you're going to take lots of notes. You might capture some video, you might shoot some photos, and that's going to form at least one bucket of raw data that results from your connection with users. The second approach to connect with users, is that you will actively intervene. That is, you will actively conduct interviews with users in your target groups or target segments. As I indicated earlier, one-on-one interviews are the most efficient. That's the way you're going to wanbt to do it. If you want to cover as much ground as possible, Even if you work in a team in designing an artifact, you're going to find that one-on-one interviews will let you cover more users with less effort, than if you do it as a team. Again, you very much want to interview users in their use environment. You're going to learn things about their challenges and needs that you wouldn't learn by e-mail or on the phone or in an office. So, you really want to find them where they work or where they use the artifact or where the problem exists in order to best understand their needs. You really don't need structured interviews for to connect with users, you can ask some open ended questions like, show me how you do x? Why did you choose x? What you like about x? What don't you like about x? What suggestions for improvements do you have? What issues do you consider when choosing an x? And, those simple questions will be sufficient usually to start a dialogue and to reveal quite a bit of information about their needs. Now, I am going to show you a little bit of video here where I just interviewed my son who, was a good sport but he's a teenager, and he was a little bit reluctant to do this. But I just had him, just in one take, without giving him any prompts, I just asked him to, to prepare a couple of bowls of ice cream for him and his friend. And I want you to watch this video and observe what it is you can learn about the user needs for an ice cream scoop just by watching this video. Alright. Would you, would you scoop a couple of bowls of ice cream for you and your friend? So hopefully, you observed in that first clip that he can't find the scoop. He doesn't know where it is. Now, despite this being his kitchen, he can't find the scoop. He doesn't know where it is. He ends up trying three drawers before he finds the scoop. So, those are the kinds of, of detailed observations that you're going to make based on just few seconds of video. Okay, let's watch a little more. Why did you wash that? So, it's hotter. So, it's hotter? Yeah. Why was that a concern? So, it scoops through the ice cream more easy, easier. So, it scoops through the ice cream more easily? Correct. Now, Look at what he's doing. He's running the scoop under hot water, and when I ask him why, he says, he's trying to heat it up because it'll help him cut or form the ice cream better. And again, those are some detailed observations that you could make in just a few seconds about a peculiar behavior. You wouldn't get that behavior if you had just done an interview or asked him questions by e-mail. It was only by observing him in the use environment that we we're able to get those insights. Okay, let's keep watching. Why are you holding it that way? More leverage. More leverage? Correct. My thumb is stronger than my four fingers. Mm-hm. Why are you doing that? To heat up the, to heat up the scooper. So, it won't stick? So, it melts the ice cream. Cuz it's a higher, the, the metal head of the scoop conducts heat, and it becomes warmer so it's, a lot warmer, and it cuts through the ice cream easier. Okay. Anything you find frustrating about that experience? The water shoots in my face or over me when I. What about the way you're scooping it? Oh, that doesn't seem to come, I have to get in shavings and the ice cream doesn't seem to stick together very easily. But, and, and it's hard, it requires a lot of energy and force. Now, it's pretty interesting to look at how he's actually using the scoop. He's gripping the scoop like a plunger. Kind of like he's, he's plunging with his pinky facing downward. He's talking about his ability to apply leverage with his thumb. See how he's using the scraping action to create these curls of ice cream because he's finding the ice cream too hard. Notice that a chunk of ice cream goes flying off in the air. But quite luckily, it actually lands in the bowl. And then you see he runs the hot water again, because he thinks it needs to be heated up again. Again, all of those observations are based on actually watching him use the scoop in the use environment, and those are things you wouldn't be able to observe or gather in a more structured interview format. As a result of that effort that is connecting with users both by passive observation of users in the use environment and through structured interviews, where again, you are going to go in the use environment and you're going to ask some open-ended comments. You're going to end up with a bunch of notes, and in, you're probably going to end up with two or three pages of notes for each interviewer, for each interview that you conduct. So, you might have twenty or 30 pages of notes after connecting with your users. You might also have some video and some photos. Those are really useful for documenting the process and for communicating it to others. Ideally, your notes contain some so-called verbatim comments, that is comments in the words of the user, the so-called voice of the customer that describes some of the issues they face in dealing with the artifact. Audio recording is quite unwieldy because it requires so much effort to transcribe the, the audio record. So, as a general rule, I would recommend visual data that has video and photos. But, but mostly you're going to rely, going to rely on taking notes and trying capture the verbatim comments of your
