1 00:00:07,060 --> 00:00:08,530 Good afternoon everybody. 2 00:00:08,530 --> 00:00:11,080 Thanks for choosing to spend the next half hour with me. 3 00:00:12,200 --> 00:00:17,120 And I'm gonna talk about, how to figure things out. 4 00:00:17,120 --> 00:00:20,880 How to make sense and, in my professional life there's kind of 5 00:00:20,880 --> 00:00:25,280 three times when I have to figure things out. 6 00:00:25,280 --> 00:00:27,460 Firstly, it's making sense of data. 7 00:00:28,570 --> 00:00:33,490 That can be quantitative data but in my case it's often qualitative information. 8 00:00:33,490 --> 00:00:36,150 I go out, I observe people. 9 00:00:36,150 --> 00:00:40,640 And I have to figure out from what I see, what does it mean and 10 00:00:40,640 --> 00:00:41,690 how does it actually matter. 11 00:00:43,560 --> 00:00:48,809 Secondly most of us these days work in teams and we have to figure out 12 00:00:48,809 --> 00:00:53,707 together are we on the [NOISE] thanks for kicking over my timer. 13 00:00:53,707 --> 00:00:57,038 [LAUGH] It still works. 14 00:00:57,038 --> 00:00:58,884 No troubles. 15 00:00:58,884 --> 00:01:02,860 [LAUGH] So we have to figure out together. 16 00:01:02,860 --> 00:01:05,240 Are we mean, are we meaning the same things? 17 00:01:05,240 --> 00:01:06,850 Are we agreed on the vision? 18 00:01:06,850 --> 00:01:09,040 Have we agreed how to build whatever we're building? 19 00:01:10,120 --> 00:01:14,028 And then thirdly, it's actually designing things that make sense. 20 00:01:14,028 --> 00:01:18,230 Cuz even if we understand our customers, we work really well as a team, 21 00:01:18,230 --> 00:01:21,180 we can still make mistakes and think this makes a lot of sense, 22 00:01:21,180 --> 00:01:25,020 put it in somebody's hands and realize it's not as intuitive as we were hoping. 23 00:01:26,490 --> 00:01:31,090 So I'm gonna share some of the tools and the concepts that helped me. 24 00:01:31,090 --> 00:01:34,050 It's really things that shape my practice, that shape my thinking. 25 00:01:38,020 --> 00:01:43,141 [NOISE] So first let me start with talking a bit about research. 26 00:01:43,141 --> 00:01:47,720 I do various things in my current job, my current job title is Product Architect. 27 00:01:47,720 --> 00:01:49,970 Yeah, my, my, it's kind of made up. 28 00:01:50,990 --> 00:01:53,800 But one of my roles is to go out and 29 00:01:53,800 --> 00:01:58,510 either observe something that people are already using, it can be competitors or 30 00:01:58,510 --> 00:02:03,080 it can be their existing workflow, if we're introducing new products to them. 31 00:02:03,080 --> 00:02:05,060 And see how they're we getting on. 32 00:02:05,060 --> 00:02:10,330 Sometimes it can be taking a prototype, taking a concept that we are working on, 33 00:02:10,330 --> 00:02:11,410 put it in front people. 34 00:02:12,780 --> 00:02:18,510 Now, whenever I do research I always try to combine just interviewing people but 35 00:02:18,510 --> 00:02:21,250 also showing them something at the same time. 36 00:02:21,250 --> 00:02:23,500 And whenever possible I love to go out and 37 00:02:23,500 --> 00:02:29,130 actually do this in their context to see what devices are they're actually using. 38 00:02:29,130 --> 00:02:32,920 What are their, what are their browsers of choice, what are the tools around them, 39 00:02:32,920 --> 00:02:35,620 do they still use other things that I wouldn't have considered. 40 00:02:35,620 --> 00:02:38,090 And if somebody uses their own machine or 41 00:02:38,090 --> 00:02:41,280 you're at their desk in their office, that's something that's really, 42 00:02:41,280 --> 00:02:44,450 really good to see that wouldn't get otherwise if you only use Ability Lab. 43 00:02:45,640 --> 00:02:48,690 Now I currently work with a client in healthcare. 44 00:02:48,690 --> 00:02:51,540 And in this case it's very hard to get people into a lab. 45 00:02:51,540 --> 00:02:55,200 You know, if you wanna see what's going on in a hospital you have to go and 46 00:02:55,200 --> 00:02:56,490 you just have to be flexible and 47 00:02:56,490 --> 00:02:59,750 whenever somebody has time you have to be ready to talk to them now. 48 00:03:00,960 --> 00:03:03,545 This also makes it very hard to take my team with me. 49 00:03:03,545 --> 00:03:07,930 Cuz it's hard often to get into these places just as one person that's not 50 00:03:07,930 --> 00:03:10,690 allowed to, you know, record anything or take any pictures. 51 00:03:10,690 --> 00:03:11,870 You're just relying on your notes. 52 00:03:13,170 --> 00:03:16,600 And I thought this experience that when you are presenting things to people or 53 00:03:16,600 --> 00:03:18,230 you're watching them use something, 54 00:03:18,230 --> 00:03:21,500 you can get very tactical insights really quickly. 55 00:03:21,500 --> 00:03:25,730 And people who start bringing research to an organization who has maybe relied on 56 00:03:25,730 --> 00:03:29,940 quantitative data about their customers, they often have quick wins. 57 00:03:29,940 --> 00:03:31,280 Just with calling and saying, look, 58 00:03:31,280 --> 00:03:36,310 there's a few small changes we can make that will make everything more usable. 59 00:03:36,310 --> 00:03:39,090 So this is a good strategy of getting by in. 60 00:03:39,090 --> 00:03:41,720 But very often when you are out talking to people. 61 00:03:41,720 --> 00:03:45,120 You observe these kind of bigger issues that are going on and 62 00:03:45,120 --> 00:03:47,400 sometimes it's not that easy to fix. 63 00:03:47,400 --> 00:03:49,760 Sometimes the way that your product is set up and 64 00:03:49,760 --> 00:03:53,290 that can go all the way down to your back end of your data structure. 65 00:03:53,290 --> 00:03:56,860 Maybe there's just some fundamental concepts that are clear to you and 66 00:03:56,860 --> 00:03:59,570 your organization but people just don't really get it. 67 00:04:00,690 --> 00:04:04,500 And a great tool to actually get that across, especially if you and 68 00:04:04,500 --> 00:04:08,810 your team are out there, but not everybody has the great insight. 69 00:04:08,810 --> 00:04:11,780 And to actually battle for these bigger changes that can be quite hard, 70 00:04:11,780 --> 00:04:13,940 because sometimes you have to remove things and 71 00:04:13,940 --> 00:04:18,680 take your product apart a bit, is a tool called the Mental model. 72 00:04:18,680 --> 00:04:20,840 So, I don't know if you've heard about it. 73 00:04:20,840 --> 00:04:23,645 But, Indi Young wrote a really great book on the topic, 74 00:04:23,645 --> 00:04:25,570 published by Rosenfeld Media, that is, 75 00:04:25,570 --> 00:04:29,390 you know, in all details explaining to you how to do this technique. 76 00:04:29,390 --> 00:04:33,500 But the concept, the term mental model has been around before Indi used it for 77 00:04:33,500 --> 00:04:34,380 user experiences. 78 00:04:35,450 --> 00:04:39,950 And the Mental model is, it's an explanation of somebody's thought process, 79 00:04:39,950 --> 00:04:41,690 about how something works in the world. 80 00:04:43,560 --> 00:04:46,560 And it's a representation of the surrounding world. 81 00:04:46,560 --> 00:04:51,390 Or the relationships between it and the rarest parts in it. 82 00:04:51,390 --> 00:04:55,860 And a person's intuitive perception about the impact that his or 83 00:04:55,860 --> 00:04:58,250 her own act, actions will have. 84 00:04:58,250 --> 00:05:00,510 And the word intuitive is really, really important here. 85 00:05:01,540 --> 00:05:05,700 If we weren't able to categorize the world and build up models in our heads about how 86 00:05:05,700 --> 00:05:09,610 something works, we would suffer from cognitive overload all the time, and 87 00:05:09,610 --> 00:05:13,630 wouldn't be able to make decisions, or solve problems at all. 88 00:05:15,210 --> 00:05:18,404 So Mental models can really help shape behavior and 89 00:05:18,404 --> 00:05:22,050 be set an approach to solving problems and to doing tasks. 90 00:05:23,120 --> 00:05:26,640 So, it's important to see what are existing mental models people have. 91 00:05:26,640 --> 00:05:29,260 These can be based on competitor products, 92 00:05:29,260 --> 00:05:33,760 on their training that they've received, on existing best practices out there. 93 00:05:33,760 --> 00:05:35,950 And sometimes you want to design for this Mental model. 94 00:05:37,350 --> 00:05:42,860 However, sometimes you want to get people, help people form a new mental model, 95 00:05:42,860 --> 00:05:45,630 break with what they know and give them something different, something better. 96 00:05:47,830 --> 00:05:52,588 This is how it normally works when you put a mental model together, writing a lot of 97 00:05:52,588 --> 00:05:57,660 post-its and jotting down what's the task that people actually do? 98 00:05:57,660 --> 00:05:58,840 What's their work flow? 99 00:05:58,840 --> 00:06:00,060 What's their decision process? 100 00:06:00,060 --> 00:06:01,440 How are they going about things? 101 00:06:02,890 --> 00:06:04,980 And you know then what's happening. 102 00:06:04,980 --> 00:06:09,420 But a really important thing is to pay attention to, what are people's own words? 103 00:06:09,420 --> 00:06:11,980 How are they describing what they're doing? 104 00:06:11,980 --> 00:06:15,410 Because it's really the language that people use that often expresses the model 105 00:06:15,410 --> 00:06:18,460 that they have in their heads, and that can differ quite a lot 106 00:06:18,460 --> 00:06:22,900 from the product that you have in your head and the language that you are using. 107 00:06:22,900 --> 00:06:25,710 And this is great to do with everybody who goes out and does research but I 108 00:06:25,710 --> 00:06:30,330 think also involving customer support with sales people is a really great idea here. 109 00:06:30,330 --> 00:06:34,185 So in the healthcare sector, the sales guys that I'm working with are all trained 110 00:06:34,185 --> 00:06:37,810 healthcare professionals and they can tell me a lot about my assumptions about 111 00:06:37,810 --> 00:06:41,390 why are people deserving, like why are they doing this in a certain way. 112 00:06:41,390 --> 00:06:42,090 What's going on here? 113 00:06:42,090 --> 00:06:43,710 They're always choosing this one approach. 114 00:06:43,710 --> 00:06:46,960 And they can tell me, oh, because that's how you train in med school. 115 00:06:46,960 --> 00:06:50,147 That's what it is, and I wouldn't have an idea about that. 116 00:06:50,147 --> 00:06:53,249 Deliverables of mental models, deliverables, like, 117 00:06:53,249 --> 00:06:56,300 models, look like this, sometimes. 118 00:06:56,300 --> 00:06:57,200 Here's another one. 119 00:06:59,170 --> 00:07:00,890 And here's another one. 120 00:07:00,890 --> 00:07:03,260 What all of them have in common as you can see here. 121 00:07:03,260 --> 00:07:05,390 It's basically mapping out the tasks and 122 00:07:05,390 --> 00:07:08,680 this can be tasks where your product is being used. 123 00:07:08,680 --> 00:07:11,240 But it can also be before and after. 124 00:07:11,240 --> 00:07:14,850 And the green part is all the actions that people take. 125 00:07:14,850 --> 00:07:16,470 What they say. 126 00:07:16,470 --> 00:07:19,690 What problems they have and their language about it. 127 00:07:19,690 --> 00:07:24,620 And then the blue parts here is mapping your product against it and seeing what's 128 00:07:24,620 --> 00:07:28,200 the language we use, what are the features we are providing or we want to provide. 129 00:07:28,200 --> 00:07:30,360 And you can also spot gaps in here. 130 00:07:31,930 --> 00:07:36,040 So I find it quite useful as an exercise to do with people to stimulate these 131 00:07:36,040 --> 00:07:38,800 discussions about what are the bigger things that are going on and 132 00:07:38,800 --> 00:07:41,050 making sense of these deep problems. 133 00:07:41,050 --> 00:07:43,750 That sometimes require much bigger changes in 134 00:07:43,750 --> 00:07:46,400 a product than you were willing to make in the beginning. 135 00:07:48,620 --> 00:07:52,760 And it's really about figuring out, why are people doing something. 136 00:07:52,760 --> 00:07:56,600 And helping you figure out what to do next. 137 00:07:56,600 --> 00:08:00,720 So I found this is a good technique to help everybody agree on these kind of 138 00:08:00,720 --> 00:08:01,510 bigger issues, and 139 00:08:01,510 --> 00:08:05,530 as a person going out and getting feedback from customers has helped me to, 140 00:08:05,530 --> 00:08:10,010 just over a period of time that I am lucky to spend with the same group of users. 141 00:08:10,010 --> 00:08:15,410 To just understand is there any big mismatch between what 142 00:08:15,410 --> 00:08:19,120 they have in their heads and how we've designed, how we've structured everything. 143 00:08:21,310 --> 00:08:25,570 So that's one of the things where I feel I have to make sense of things I see and 144 00:08:25,570 --> 00:08:27,280 share it with others, and align, 145 00:08:27,280 --> 00:08:29,520 and this is one of the tools that I found really useful. 146 00:08:32,380 --> 00:08:38,150 Secondly, I've been lucky in the last few years to mainly work on new products. 147 00:08:38,150 --> 00:08:41,470 Something that hasn't existed before. 148 00:08:41,470 --> 00:08:44,480 And I've been lucky to always work in teams that have all the skills we 149 00:08:44,480 --> 00:08:49,560 needed and this is when I realized I love working with back end developers. 150 00:08:49,560 --> 00:08:51,550 And I love working with data things and 151 00:08:51,550 --> 00:08:54,070 especially when it's, how are we gonna get the data? 152 00:08:54,070 --> 00:08:55,620 How's it gonna look in the database? 153 00:08:55,620 --> 00:08:57,570 How are we gonna structure it? 154 00:08:57,570 --> 00:08:59,860 How are we gonna collect it, where are we getting it from? 155 00:08:59,860 --> 00:09:05,330 And then how do I re-jiggle it, to actually present it to users, and 156 00:09:05,330 --> 00:09:08,280 with the intent of what we want them to take away. 157 00:09:08,280 --> 00:09:11,116 I completely love that process, and I was like ignoring some of 158 00:09:11,116 --> 00:09:15,200 the nice front end decisions, and like oh back end guys, let's do this together. 159 00:09:15,200 --> 00:09:16,270 So I really loved that, but 160 00:09:16,270 --> 00:09:21,400 I was lucky that I was there when these fundamental decisions were being made. 161 00:09:21,400 --> 00:09:23,400 I worked their large Telco before, 162 00:09:23,400 --> 00:09:27,610 and a lot of back end decisions had been made ages ago and 163 00:09:27,610 --> 00:09:31,980 when we were coming in it was like oh, not possible, not possible, not possible. 164 00:09:31,980 --> 00:09:33,330 Takes too much rework. 165 00:09:33,330 --> 00:09:35,430 So that was such a shame, such a missed opportunity. 166 00:09:37,270 --> 00:09:41,000 Because we might know about people's Mental Models and the ones that they 167 00:09:41,000 --> 00:09:45,450 existingly have, but often there's a big gap between the Mental Model, and 168 00:09:45,450 --> 00:09:48,550 the model that we've decided our system to have, our data to have. 169 00:09:48,550 --> 00:09:54,220 And this maybe sounds like it was the old and dark days where we all had to 170 00:09:54,220 --> 00:09:58,590 understand technology like very properly to be able to use anything. 171 00:09:58,590 --> 00:10:02,940 But there are a lot of people out there who still suffer from the System Model 172 00:10:02,940 --> 00:10:05,070 being exposed on the interface layer, 173 00:10:05,070 --> 00:10:09,590 preventing them from really building up an understanding about how that works. 174 00:10:09,590 --> 00:10:13,620 An example is often enterprise software business to business solutions. 175 00:10:14,730 --> 00:10:18,480 So this is, I worked on a financial services project earlier this year and 176 00:10:18,480 --> 00:10:22,460 I was going out and observing people using what 177 00:10:22,460 --> 00:10:26,400 they're using right now to help us build up a strategy of what we wanted to build. 178 00:10:26,400 --> 00:10:29,030 And this is a screenshot of one of the services. 179 00:10:29,030 --> 00:10:30,720 It's not, yeah. 180 00:10:30,720 --> 00:10:32,750 It's not that great to see it here but 181 00:10:32,750 --> 00:10:35,900 it gives you an idea of this software doesn't look very great. 182 00:10:36,900 --> 00:10:39,490 But people had lots of issues with this because it 183 00:10:39,490 --> 00:10:43,760 just didn't match their expectations from how they were using the web. 184 00:10:43,760 --> 00:10:47,430 So for example when you were logging in it didn't show you your information that you 185 00:10:47,430 --> 00:10:50,410 wanted to see and for a lot of people that wasn't very much. 186 00:10:50,410 --> 00:10:51,700 You had to search. 187 00:10:51,700 --> 00:10:54,280 People like, I don't really know what to search for. 188 00:10:54,280 --> 00:10:55,510 Now back in the olden days, 189 00:10:55,510 --> 00:11:00,560 performing an empty search often tricks the database to turn, return all results. 190 00:11:00,560 --> 00:11:02,230 A few people had figured that out and 191 00:11:02,230 --> 00:11:07,676 were so happy, but others really struggled to even get started with the system. 192 00:11:07,676 --> 00:11:08,900 There were a few other things like, 193 00:11:08,900 --> 00:11:12,360 to get your account details, you couldn't just get your account details. 194 00:11:12,360 --> 00:11:14,260 You had to run a report. 195 00:11:14,260 --> 00:11:16,240 For every single thing, you had to run a report, 196 00:11:16,240 --> 00:11:21,070 which makes sense, because that's how things were built to work at the back end. 197 00:11:21,070 --> 00:11:26,060 But a lot of these people are very young administrators who use technology in 198 00:11:26,060 --> 00:11:27,360 a very different way. 199 00:11:27,360 --> 00:11:30,330 And they just completely lacked these mental models about how to 200 00:11:30,330 --> 00:11:33,580 trick an older system into giving them what they wanted. 201 00:11:36,730 --> 00:11:41,000 And to bridge these two, there's a way of doing that and 202 00:11:41,000 --> 00:11:43,650 it's called a Conceptual Model. 203 00:11:43,650 --> 00:11:47,220 So a Conceptual Model is really an expression of what we 204 00:11:47,220 --> 00:11:50,230 want users to be able to do with our product. 205 00:11:51,350 --> 00:11:52,990 And it can be quite abstract. 206 00:11:52,990 --> 00:11:58,240 It's basically deciding, what are all the things that we want people to do. 207 00:11:58,240 --> 00:12:01,410 What are the tasks we want them to do, what are all the entities? 208 00:12:01,410 --> 00:12:04,780 What are the elements, the objects in there, and how can I interact with them? 209 00:12:05,940 --> 00:12:08,570 And you could say that if you write up a backlog, 210 00:12:08,570 --> 00:12:13,710 if you write a list of features, this can be a conceptual model. 211 00:12:13,710 --> 00:12:17,050 But I think we have to go a bit further of defining this together. 212 00:12:19,310 --> 00:12:21,660 Because conceptual models will matter in the future. 213 00:12:23,770 --> 00:12:27,960 Right now it's fairly easy to have an understanding about how your house works. 214 00:12:27,960 --> 00:12:31,720 If the electricity goes out, I know exactly how to fix it. 215 00:12:31,720 --> 00:12:34,272 But if you're thinking about the future of connected homes. 216 00:12:34,272 --> 00:12:37,703 The mental models we have right now about how our house, 217 00:12:37,703 --> 00:12:41,444 how our house works, will be very, very different going forward. 218 00:12:41,444 --> 00:12:43,270 And if everything's connected, 219 00:12:43,270 --> 00:12:46,680 connected to servers talking to the internet talking to our phones, 220 00:12:46,680 --> 00:12:52,140 if something breaks we have to figure out how to fix it as users of this technology. 221 00:12:52,140 --> 00:12:56,210 So I think while we don't, not many of us work on enterprise software and 222 00:12:56,210 --> 00:13:00,690 like these dark days of designing things, but this is a new frontier for me. 223 00:13:00,690 --> 00:13:04,270 And I think this is almost the next one we have, where we have to make 224 00:13:04,270 --> 00:13:10,460 sure we are teaching people correctly about how everything works. 225 00:13:10,460 --> 00:13:13,170 And this is when the conceptual model I think, is really important. 226 00:13:14,470 --> 00:13:16,480 The banking software that I showed you. 227 00:13:16,480 --> 00:13:20,720 A lot of users said, only knew how to do things in one way. 228 00:13:20,720 --> 00:13:22,350 And they were not clicking anywhere else. 229 00:13:23,480 --> 00:13:25,900 They had what we call, procedural knowledge. 230 00:13:25,900 --> 00:13:29,780 So if you are going to a city where your friend lives, and you've never been there, 231 00:13:29,780 --> 00:13:33,750 but you know how to get to their house, you will probably just walk the one way, 232 00:13:33,750 --> 00:13:35,760 because if you turn a corner you might even be lost. 233 00:13:35,760 --> 00:13:36,940 You don't really know what's going on. 234 00:13:38,050 --> 00:13:40,755 However, if you have a concept of how the city's structured, 235 00:13:40,755 --> 00:13:44,570 maybe it as landmarks or river that helps you sort out where you are. 236 00:13:44,570 --> 00:13:46,380 If you know where north and south, west and 237 00:13:46,380 --> 00:13:48,630 east is, you have something called survey knowledge. 238 00:13:48,630 --> 00:13:52,930 And when you have survey knowledge, you're more likely to explore but you understand, 239 00:13:52,930 --> 00:13:56,790 in this case, how a city's connected and where you have to walk to get somewhere, 240 00:13:56,790 --> 00:13:58,458 approximately, and you can figure it out. 241 00:13:58,458 --> 00:14:01,770 And the conceptual mode is really about giving the people 242 00:14:01,770 --> 00:14:05,610 the survey knowledge to help them understand how something works. 243 00:14:06,700 --> 00:14:10,830 And with new things like this or even just, you know, it can be mind 244 00:14:10,830 --> 00:14:15,770 boggling that something actually works on your, on a website in your browser. 245 00:14:15,770 --> 00:14:18,860 But you can also have a mobile phone if everything talks to each other. 246 00:14:18,860 --> 00:14:21,990 There's so many things that we are working on now that are like already taken 247 00:14:21,990 --> 00:14:22,970 for granted. 248 00:14:22,970 --> 00:14:25,330 But it's our job to help people really understand those. 249 00:14:27,790 --> 00:14:32,710 Now when collaborating together on defining things early on, as I mentioned, 250 00:14:32,710 --> 00:14:37,860 I was very lucky to work with a great team that every, we had all the kind of 251 00:14:37,860 --> 00:14:40,790 business design and tech skills that we needed to form something new. 252 00:14:42,700 --> 00:14:46,710 But it's still often hard to then agree what we're gonna build together. 253 00:14:47,980 --> 00:14:50,330 A lot of the start up teams that I've mentored, 254 00:14:50,330 --> 00:14:52,310 one of the fun things that you get to do. 255 00:14:52,310 --> 00:14:54,490 You get invited to meetings, but you don't have to do anything. 256 00:14:54,490 --> 00:14:55,450 You just have to observe. 257 00:14:57,090 --> 00:15:00,590 And I've often seen them jumping back and forth between, 258 00:15:00,590 --> 00:15:04,750 you know, what's our current product and what we want to fix in the short term, but 259 00:15:04,750 --> 00:15:07,520 here's the bigger roadmap, and here's all the things that are broken. 260 00:15:07,520 --> 00:15:11,500 You know, seven, eight people looking at visual designs, but 261 00:15:11,500 --> 00:15:16,570 discussing everything from all different layers across what makes up the product. 262 00:15:16,570 --> 00:15:20,850 And it can be quite hard to then figure out where you wanna go and 263 00:15:20,850 --> 00:15:22,005 if we're actually agreeing. 264 00:15:22,005 --> 00:15:27,610 Cuz I think collaboration problems arise when teams lack the tools to agree meaning 265 00:15:27,610 --> 00:15:32,400 and structure, and that is key to make sense of what we're building together. 266 00:15:32,400 --> 00:15:35,030 We have to agree on what we're achieving. 267 00:15:35,030 --> 00:15:35,780 What's the meaning? 268 00:15:35,780 --> 00:15:36,720 What's our intent? 269 00:15:36,720 --> 00:15:38,470 What do we want to get across? 270 00:15:38,470 --> 00:15:41,470 And what's the underlying structure that holds everything together? 271 00:15:42,470 --> 00:15:44,490 And I think structure really matters. 272 00:15:44,490 --> 00:15:48,510 In Megan's talk she talked about how her clients were often bad clients when 273 00:15:48,510 --> 00:15:51,500 they hadn't figure out what their intent is. 274 00:15:51,500 --> 00:15:54,900 And when there wasn't a fundamental structure in the product. 275 00:15:54,900 --> 00:15:59,190 What Anna talked about just now about the console browsers it made me think, no 276 00:15:59,190 --> 00:16:04,030 matter what you're showing on both device, if you have an underlying structure of 277 00:16:04,030 --> 00:16:08,450 content and of how things are connected in your product, that should be there. 278 00:16:08,450 --> 00:16:12,240 That will work no matter what a user is using to access your information. 279 00:16:14,760 --> 00:16:20,570 And the tool that helps me to pick the right technique, for 280 00:16:20,570 --> 00:16:24,140 me to explain what I mean and share with others. 281 00:16:24,140 --> 00:16:27,710 And agree, there's common understanding of what we're actually doing. 282 00:16:27,710 --> 00:16:29,550 For me, that's really information architecture. 283 00:16:30,800 --> 00:16:33,260 And I'm gonna break it, information architecture, down into kind of, 284 00:16:33,260 --> 00:16:35,200 three components of it. 285 00:16:36,550 --> 00:16:38,080 And the first one is Ontology. 286 00:16:39,160 --> 00:16:42,000 Do you know what you mean when you say what you say? 287 00:16:43,340 --> 00:16:48,640 So this is really about figuring out what is the thing we're building made of and 288 00:16:48,640 --> 00:16:49,630 what are we gonna call it. 289 00:16:50,780 --> 00:16:51,410 Think about it. 290 00:16:51,410 --> 00:16:55,310 A lot of the things we make these days are actually made out of words. 291 00:16:55,310 --> 00:16:57,010 Content and words will be there. 292 00:16:57,010 --> 00:17:01,460 The things we, how we label things, how we categorize things is often very, 293 00:17:01,460 --> 00:17:03,050 very important. 294 00:17:03,050 --> 00:17:05,770 And if we pick a certain language, 295 00:17:05,770 --> 00:17:10,260 that often has a certain meaning that changes how a product feels to others. 296 00:17:11,330 --> 00:17:14,150 There's tools to define ontologies. 297 00:17:14,150 --> 00:17:16,570 One of them is a controlled vocabulary, 298 00:17:16,570 --> 00:17:20,410 where you decide these are all the words that we are gonna use. 299 00:17:20,410 --> 00:17:22,870 Here's all the other stuff that we don't want to have in it. 300 00:17:24,130 --> 00:17:25,790 I've seen this used in design guidelines. 301 00:17:26,910 --> 00:17:30,640 Very often, good content editors will have something like this, 302 00:17:30,640 --> 00:17:32,750 where they say, these are all the things we want to use. 303 00:17:32,750 --> 00:17:34,610 This is how we're going to talk about it. 304 00:17:34,610 --> 00:17:38,550 But I've also just seen it in teams when they collaborate and agree early on. 305 00:17:38,550 --> 00:17:40,880 This is how we're gonna call something for these reasons, and 306 00:17:40,880 --> 00:17:42,840 we all agree what it means. 307 00:17:42,840 --> 00:17:45,180 So it's not really a tool for me. 308 00:17:45,180 --> 00:17:47,530 It's more something that I'm just always aware of, 309 00:17:47,530 --> 00:17:50,870 just think about what decisions are we making just about language? 310 00:17:50,870 --> 00:17:54,060 And then secondly it's about Taxonomy. 311 00:17:54,060 --> 00:17:56,031 This is often what people think of when they, 312 00:17:56,031 --> 00:17:58,130 when I say, information architecture. 313 00:17:58,130 --> 00:18:01,890 Have you provided logical structures that bring meaning? 314 00:18:01,890 --> 00:18:03,590 How are you organizing things? 315 00:18:03,590 --> 00:18:09,380 You can have sitemaps, you can organize things in facets, but 316 00:18:09,380 --> 00:18:13,720 just thinking about, how does everything hang together and how does it make sense. 317 00:18:16,730 --> 00:18:21,130 And because we have to make sure while our products are constantly evolving, 318 00:18:21,130 --> 00:18:24,540 if you start content first, you have to figure out, is there a place for 319 00:18:24,540 --> 00:18:27,430 content across what we're building and will there be, 320 00:18:27,430 --> 00:18:29,690 where is gonna be spaces for content in the future. 321 00:18:29,690 --> 00:18:33,570 And the thing that brings it all together is Choreography. 322 00:18:33,570 --> 00:18:38,620 So, How is meaning affected across various channels and over time and through usage? 323 00:18:38,620 --> 00:18:42,520 And this is why we have things like flowcharts, swimlane diagrams but 324 00:18:42,520 --> 00:18:44,810 for me, this is merely prototyping. 325 00:18:44,810 --> 00:18:48,570 To figure out, okay, we've defined what's part of what we're building. 326 00:18:48,570 --> 00:18:52,560 How is it all structured and connected, but now, how do I interact with it? 327 00:18:52,560 --> 00:18:54,420 What are all the user journeys that we're gonna have? 328 00:18:55,510 --> 00:19:00,370 My workflow at the moment is a lot taking, sketching something and 329 00:19:00,370 --> 00:19:03,898 putting it into POP, prototyping on paper, on the phone. 330 00:19:03,898 --> 00:19:07,467 I mock something up in whatever tool I'm comfortable using and 331 00:19:07,467 --> 00:19:10,100 I put it in tools like Flint or Invision. 332 00:19:10,100 --> 00:19:13,370 And I do this just for myself as a way of thinking about it, but 333 00:19:13,370 --> 00:19:16,950 also to share it with my team and see is it all holding together. 334 00:19:16,950 --> 00:19:20,230 Because there's so many things you're not necessarily considering and 335 00:19:20,230 --> 00:19:23,380 the structure that feels really good if you're considering, 336 00:19:23,380 --> 00:19:25,300 when is somebody gonna access this? 337 00:19:25,300 --> 00:19:26,180 How are they gonna do it? 338 00:19:26,180 --> 00:19:28,010 What've, what've they done before? 339 00:19:28,010 --> 00:19:29,590 That element of choreography and 340 00:19:29,590 --> 00:19:32,830 how users are going to interact is something really important. 341 00:19:32,830 --> 00:19:34,410 And the earlier you model this and 342 00:19:34,410 --> 00:19:37,690 you prototype this, the easier it will be for you to understand. 343 00:19:37,690 --> 00:19:40,600 Have we thought everything through, does it all make sense. 344 00:19:42,010 --> 00:19:46,980 So these three things together make up the wider field of information architecture. 345 00:19:46,980 --> 00:19:52,280 If you're interested Dan Klyn, Abby Covert are people who are great looking up deeper 346 00:19:52,280 --> 00:19:56,200 definitions about this and the guy called Peter Morville has just written a book 347 00:19:56,200 --> 00:20:00,580 called Intertwingled, that goes a lot into the topics I'm talking about in this talk. 348 00:20:02,910 --> 00:20:04,810 So these are kind of my tools. 349 00:20:04,810 --> 00:20:08,290 Information architecture helps me to be mindful about the decisions we're making. 350 00:20:08,290 --> 00:20:13,630 But it also gives me a range of different tools, diagrams, things that I can use. 351 00:20:13,630 --> 00:20:18,140 So go out and use diagrams, use prototypes to think for yourself. 352 00:20:18,140 --> 00:20:21,280 But more importantly, communicate with each other. 353 00:20:21,280 --> 00:20:26,560 And it's not that prototypes are deliverables and mental models and 354 00:20:26,560 --> 00:20:29,840 drawing up are kind of things that are sorted now. 355 00:20:29,840 --> 00:20:31,540 It's just tools to help us think and 356 00:20:31,540 --> 00:20:34,330 to see are we on the same picture, are we all agreeing. 357 00:20:38,910 --> 00:20:40,790 But I often still get it wrong, 358 00:20:40,790 --> 00:20:44,240 making sense of things is really hard working with other people. 359 00:20:44,240 --> 00:20:45,720 Is really, really hard. 360 00:20:46,760 --> 00:20:49,770 This is actually not an ant. 361 00:20:49,770 --> 00:20:51,610 I struggle to pronounce the word. 362 00:20:51,610 --> 00:20:54,430 But this is a spider disguising as one. 363 00:20:54,430 --> 00:20:57,610 So often things are not quite as they seem. 364 00:20:57,610 --> 00:21:02,640 But, just by using some of the tools and just thinking about things that I've, 365 00:21:02,640 --> 00:21:04,930 just a few concepts I've just explained to you. 366 00:21:04,930 --> 00:21:07,460 That helps me to figure out things and 367 00:21:07,460 --> 00:21:09,230 hopefully designing things that make sense. 368 00:21:10,230 --> 00:21:15,570 Now, I looked up sensemaking as a term because there's a big field of people who 369 00:21:15,570 --> 00:21:19,640 are actually talking about complexity and systems thinking and sensemaking. 370 00:21:19,640 --> 00:21:24,627 And what was interesting to me is that sensemaking is taught in business schools, 371 00:21:24,627 --> 00:21:29,352 so what I'm gonna show you is from a book by a lady from the MIT. 372 00:21:29,352 --> 00:21:35,380 And sensemaking is there taught as a key leadership capability in MBA programs. 373 00:21:35,380 --> 00:21:40,290 And I was interested what the sensemaking mean in this kind of business context and 374 00:21:40,290 --> 00:21:44,390 does it apply to my understanding of, you know, figuring things out. 375 00:21:44,390 --> 00:21:46,570 Collecting data, modeling it, 376 00:21:46,570 --> 00:21:50,740 expressing it, prototyping it, collaborating, sharing it with others. 377 00:21:50,740 --> 00:21:54,370 And then experimenting and putting things out in the world to get more feedback, 378 00:21:54,370 --> 00:21:57,320 is that the process that these business people mean when they talk about it. 379 00:21:58,980 --> 00:22:03,880 So, they define sensemaking as it refers to how we structure the unknown, so 380 00:22:03,880 --> 00:22:05,480 as we are gonna be able to act in it. 381 00:22:05,480 --> 00:22:10,200 And sensemaking involves coming up with a plausible understanding, 382 00:22:10,200 --> 00:22:12,570 a map of a shifting of a changing world. 383 00:22:13,980 --> 00:22:18,980 When in testing this map with others through data collection, through action, 384 00:22:18,980 --> 00:22:19,860 through conversation. 385 00:22:20,920 --> 00:22:25,370 And then we refine it or abandon it depending on how credible it is. 386 00:22:27,190 --> 00:22:29,990 So this is a lot about modeling, making maps, 387 00:22:29,990 --> 00:22:31,670 collaborating, sharing the experiment. 388 00:22:31,670 --> 00:22:32,950 I found it very, very similar. 389 00:22:34,650 --> 00:22:38,060 And I think it matters for us because I'm gonna show you a quote from 390 00:22:38,060 --> 00:22:43,080 Simon Collison from 2009 where he said, what we build is rarely finished. 391 00:22:43,080 --> 00:22:47,357 We build systems that flex and grow with the client, the business, 392 00:22:47,357 --> 00:22:52,970 the organization, the community, and the availability of new devices. 393 00:22:54,110 --> 00:22:57,070 So, this was in 2009, five years ago. 394 00:22:57,070 --> 00:22:59,470 I think it's even more important now. 395 00:22:59,470 --> 00:23:03,730 I find in my job I'm often the one that, 396 00:23:03,730 --> 00:23:07,920 especially when you build new products, it is a responsibility to client's business. 397 00:23:08,960 --> 00:23:11,730 You're coming in, you're shaping a strategy. 398 00:23:11,730 --> 00:23:15,210 And a lot of it is also about how we work as an organization, 399 00:23:15,210 --> 00:23:16,980 how the client works as an organization. 400 00:23:18,190 --> 00:23:24,510 And I feel that it, the more we work in, on products that are part 401 00:23:24,510 --> 00:23:29,440 of an ecosystem of devices of other products if we work on a bigger service. 402 00:23:29,440 --> 00:23:31,920 We have to think more and more and more about this complexity. 403 00:23:34,930 --> 00:23:39,560 And I looked a little bit into, how can I unravel this complexity and 404 00:23:39,560 --> 00:23:43,490 just help figure out this, often sometimes quite challenging situations. 405 00:23:44,820 --> 00:23:48,020 And there's a framework called Cynefin by a guy called, Dave Snowden, 406 00:23:48,020 --> 00:23:51,780 that if you're interested in this you can look into this for a bit more. 407 00:23:51,780 --> 00:23:56,480 But he defines situations into categories of complexity. 408 00:23:56,480 --> 00:23:57,820 And the first one is Simple. 409 00:23:57,820 --> 00:24:00,870 And he says our approach to decision making when there is 410 00:24:00,870 --> 00:24:04,400 a simple problem is often cause and effect. 411 00:24:04,400 --> 00:24:06,260 We understand how to fix it. 412 00:24:06,260 --> 00:24:07,920 And we can just go and do it. 413 00:24:09,350 --> 00:24:11,630 The second one is Complicated. 414 00:24:11,630 --> 00:24:15,900 So if something's really complicated you maybe need more time to figure it out. 415 00:24:15,900 --> 00:24:19,640 Maybe you realize you lack the expertise to do this. 416 00:24:19,640 --> 00:24:23,790 So your strategy to resolve it is to go out build up your expertise or bring 417 00:24:23,790 --> 00:24:27,460 somebody else in who can help you figure this out, or you give yourself more time. 418 00:24:29,070 --> 00:24:32,380 But I find we often operate in very complex situations solving 419 00:24:32,380 --> 00:24:34,070 complex problems. 420 00:24:34,070 --> 00:24:35,490 When you're putting out a new product. 421 00:24:35,490 --> 00:24:37,170 When you're making changes to something, 422 00:24:37,170 --> 00:24:40,700 you don't know what's going to happen very often. 423 00:24:40,700 --> 00:24:42,600 And that's because we work with people. 424 00:24:42,600 --> 00:24:44,170 We design for people. 425 00:24:44,170 --> 00:24:48,480 And people make things complex because you can never predict their behavior. 426 00:24:48,480 --> 00:24:50,070 You can't predict how your client will behave. 427 00:24:50,070 --> 00:24:52,840 You can't predict what the users are gonna do. 428 00:24:52,840 --> 00:24:54,100 Yes, you can go and test and 429 00:24:54,100 --> 00:24:57,500 prototype it, but this is the only way to find out what you're gonna do. 430 00:24:57,500 --> 00:24:59,600 You have to run experiments. 431 00:24:59,600 --> 00:25:01,270 So for complex situations, 432 00:25:01,270 --> 00:25:04,330 running experiments is the best way to getting more information and 433 00:25:04,330 --> 00:25:07,830 help you with your decision, and make sense of what's going on. 434 00:25:07,830 --> 00:25:09,300 There's a fourth category. 435 00:25:09,300 --> 00:25:10,470 This one is called Chaos. 436 00:25:10,470 --> 00:25:13,460 There's some very philosophical discussions about chaos. 437 00:25:13,460 --> 00:25:17,690 But in a simple way, if there's chaos, you often have to fix things quickly, and 438 00:25:17,690 --> 00:25:18,710 just act really quickly. 439 00:25:22,030 --> 00:25:24,930 There's a few sensemaking principles that I have found in 440 00:25:24,930 --> 00:25:26,270 some of these business books. 441 00:25:26,270 --> 00:25:31,000 And I feel they're very much aligned with the process that we all already use. 442 00:25:32,100 --> 00:25:36,770 So, collect many types of data from different sources and bring them together. 443 00:25:36,770 --> 00:25:40,700 For me, I always advocate to get quantitative data about your customers, 444 00:25:40,700 --> 00:25:43,690 but also get qualitative data and go out there and 445 00:25:43,690 --> 00:25:45,810 combine different techniques to get data. 446 00:25:48,140 --> 00:25:50,170 Collaborate with others. 447 00:25:50,170 --> 00:25:54,500 And I wish there was more, especially in agency and consulting land. 448 00:25:54,500 --> 00:25:58,740 I wish more people would put full teams on projects from the beginning onwards. 449 00:25:58,740 --> 00:26:02,440 To allow us to have this collaboration where the people are defining the data 450 00:26:02,440 --> 00:26:06,540 structures, how the system works to collaborate really quickly with the people 451 00:26:06,540 --> 00:26:10,530 who will be defining, define the kind of layer it uses will interact with and 452 00:26:10,530 --> 00:26:13,290 bring those two together from the beginning onwards. 453 00:26:13,290 --> 00:26:16,420 So everybody has a shared understanding of who the users are and 454 00:26:16,420 --> 00:26:18,040 what their mental model is. 455 00:26:18,040 --> 00:26:22,710 And about what the technology we have chosen whether the opportunities and 456 00:26:22,710 --> 00:26:26,030 constraints, and how can we make sure those two are better aligned. 457 00:26:29,180 --> 00:26:31,580 And I can't stress enough to model and prototype. 458 00:26:31,580 --> 00:26:36,020 And that can be anything from making a prototype in 459 00:26:36,020 --> 00:26:40,230 whatever medium you're comfortable with, just drawing, drawing a diagram of your 460 00:26:40,230 --> 00:26:42,820 backend architecture on a napkin and sharing it with each other. 461 00:26:44,390 --> 00:26:48,350 I find that sometimes there is a lot of obsession about making these things and 462 00:26:48,350 --> 00:26:50,700 getting things out there really, really quickly. 463 00:26:50,700 --> 00:26:55,450 But agreeing together, what are we gonna do, how does it all connected? 464 00:26:55,450 --> 00:26:58,470 What's the structure that binds everything? 465 00:26:58,470 --> 00:27:00,120 For me, that is really, really important. 466 00:27:00,120 --> 00:27:03,970 And I see that as part of making, and as a key thing to help us collaborate and 467 00:27:03,970 --> 00:27:07,700 not as a, as a waste of time that y stops us from actually getting things out there. 468 00:27:10,620 --> 00:27:12,320 And then finally we learn from experiments. 469 00:27:13,900 --> 00:27:16,150 Again, this is a list from a business book. 470 00:27:16,150 --> 00:27:18,520 This is not a summary of my talk. 471 00:27:18,520 --> 00:27:21,960 But I think that it is really, really interesting that this way of 472 00:27:21,960 --> 00:27:25,635 working that feels so natural to lots of us. 473 00:27:25,635 --> 00:27:28,400 That books like The Lean Startup that was mentioned earlier today by Chris that 474 00:27:28,400 --> 00:27:31,300 they're already talking about, it is something that people who are in 475 00:27:31,300 --> 00:27:35,250 leadership roles in organizations are being taught to help fix 476 00:27:35,250 --> 00:27:39,640 organizational team problems and make the big decisions. 477 00:27:39,640 --> 00:27:42,500 You know, how to spend your next multimillion dollars. 478 00:27:42,500 --> 00:27:43,570 What are you gonna do it with? 479 00:27:43,570 --> 00:27:45,660 You know, how, how are you going to stack the whole department? 480 00:27:45,660 --> 00:27:46,870 What are you going to do. 481 00:27:46,870 --> 00:27:50,610 So they are trying to apply a similar kind of process to figure things out. 482 00:27:50,610 --> 00:27:55,460 And, I am going to bring this quote back again. 483 00:27:55,460 --> 00:28:01,240 I am really lucky that I have a toolbox of things that I can choose and 484 00:28:01,240 --> 00:28:04,009 by being able, having been able to observe lots of teams. 485 00:28:05,140 --> 00:28:08,140 Obviously I still make lots of mistake myselfs. 486 00:28:08,140 --> 00:28:11,840 But I've often seen what it causes if a team starts discussing something 487 00:28:11,840 --> 00:28:16,140 without the right tool, without the right collaboration technique, or 488 00:28:16,140 --> 00:28:20,040 just without a pen and pencil to draw it out together. 489 00:28:20,040 --> 00:28:22,410 So, if you are working in a team. 490 00:28:22,410 --> 00:28:25,620 You are equipped with tools to make everything better by 491 00:28:25,620 --> 00:28:29,420 just choosing the right way of expressing something that you're working on and 492 00:28:29,420 --> 00:28:31,280 sharing with others visually. 493 00:28:31,280 --> 00:28:34,120 So I encourage you to extend your tools box and 494 00:28:34,120 --> 00:28:38,481 to just let the, always let the problem that you're in, 495 00:28:38,481 --> 00:28:43,180 or the project that you're working on decide what you're gonna use but 496 00:28:43,180 --> 00:28:47,110 have a list, have, have some tools ready, have some prototyping techniques ready. 497 00:28:47,110 --> 00:28:49,430 Have some different ways of drawing up diagrams or 498 00:28:49,430 --> 00:28:51,110 collaboration tools ready for your team. 499 00:28:52,650 --> 00:28:56,870 Because sensemaking is a process and I feel that design is a really powerful tool 500 00:28:56,870 --> 00:29:01,860 to do that, and especially as the context we work in becomes more complex. 501 00:29:02,960 --> 00:29:07,060 And our clients sometimes see design as the savior. 502 00:29:07,060 --> 00:29:10,500 You're gonna come in and fix everything and make everything beautiful. 503 00:29:10,500 --> 00:29:16,080 Often you have to look into underlying flaws in structure. 504 00:29:16,080 --> 00:29:19,200 In how data's being collected and how data's be, data's being presented. 505 00:29:20,250 --> 00:29:23,670 Says a lot of underlying things that we suddenly have to deal with. 506 00:29:23,670 --> 00:29:26,380 Sometimes our role is being brought in as a strategist or 507 00:29:26,380 --> 00:29:28,310 as a designer, as a developer. 508 00:29:28,310 --> 00:29:31,370 But we often spot everything else that's going on because I 509 00:29:31,370 --> 00:29:34,270 believe we're naturally trained as thinking more holistically. 510 00:29:34,270 --> 00:29:37,590 And that's the kind of message I want you to take away, 511 00:29:37,590 --> 00:29:42,340 that you have all the tools and hopefully you can pick up some more from my talk. 512 00:29:43,610 --> 00:29:45,260 This was it for me. 513 00:29:45,260 --> 00:29:46,990 I'm around at the after party as well. 514 00:29:46,990 --> 00:29:49,090 So if anybody has any feedback or 515 00:29:49,090 --> 00:29:54,490 any further questions, let me know and I'm at @johannakoll on Twitter. 516 00:29:54,490 --> 00:29:56,830 This is my Twitter avatar as well. 517 00:29:56,830 --> 00:29:58,182 I don't look like this today. 518 00:29:58,182 --> 00:30:01,530 But you can, maybe I should have brought this to make it easier to find me. 519 00:30:01,530 --> 00:30:02,710 But you can tweet at me. 520 00:30:02,710 --> 00:30:04,460 I'll tweet back some time later today. 521 00:30:04,460 --> 00:30:07,082 And thanks very much and thanks Generate for having me. 522 00:30:07,082 --> 00:30:11,067 [APPLAUSE]