 Thanks. So, as I said, I am, that's cool. So, hi, I'm Meg Howey, and today I'm going to talk about Ask Away, which is a project that I've been working on for the last couple of years now. And it is a web app, which we designed for the general election in New Zealand last year, and designed it to help young people engage with politics. And today I'm specifically going to talk about all of the things which I did as part of my research and design process to help understand my users, but also to help understand all of the different stakeholders in the project. And I'll talk a little bit about why I think that helped make the project a success. So, next slide. Right, so this is me. I'm a designer. I'm not really a technical person. And I have almost finished my master of design at Massey University in Wellington. And this project has been my master's project this year. And last year, the year before that, it was my undergrad visual communication design project. So this is where I'm coming at it from. This is the design process. And I'm especially interested in what's sometimes called the fuzzy front end, which is all of the stuff that happens before you even get to the point where you have a concept way before you get to wireframing and interface design. So this is how you can use the design process in a really broad sense to fit everything together and figure out who's involved, how to get them on board, and where to fit your project in. And I hope that if there's anything that you guys take away from today, it's how useful design thinking can be at that early stage of a project. Because I feel like it worked. So I've been working as lucky to be working within the design and democracy project, which is a strategic research unit within the School of Design at Massey, looking at enhancing conversation and participation on social issues through design. So a lovely big broad scope there. But we've been looking specifically last year at voter turnout, youth voter turnout. And I've been working alongside On the Fence, which is another really cool web app with the same goal. So this is our problem. We have had steadily decreasing turnout in New Zealand since 1980 in elections. And that's something that's been seen all around the world in mature democracies, it seems like. The voting for the people that we want to govern us thing is sort of is losing its way a little bit. And the people who are especially affected by this are these guys, 18 to 24 year olds, often the people who aren't turning up at the polls. In 2011, somewhere between 40 and 60% of young people in New Zealand didn't vote, depending on which statistics you're going by. So a pretty serious problem for us here. And this is what we did about it. So here we go. This is Ask Away. It was built using Ruby on Rails with a bootstrap front end and we used Angular.js. And that's basically all that I'm going to say about the technical side of it. But what you can see here is a whole lot of questions from people. So anyone could log on, ask a question, vote for the ones that they thought were important. And you didn't have to log on to do that. And then compare answers. So there's more than 1,000 questions and more than 1,000 different answers on the site from various politicians. So we're really happy with the engagement that it got last year. So the approach that I took to this was user-centered design, which is basically putting all of your users' needs, wants and limitations right in the middle of everything that you're doing. And I did a little bit of user research for this talk by asking my dad what he thought user-centered design was to him. He's a sysadmin. And he said it means you have to listen to all of the boring stuff that your stakeholders have built. And I thought that was a bit sad because I really like user-centered design. And I did see a little bit of research. It's quite old now, but it was suggesting that one of the challenges that open-source software has with getting really mainstream acceptance is that it is often built for technical users. And sometimes it's hard to think about the poor struggling people who don't understand things first off. But it's so important if you want to be making something that is having big social change, if that's what you want to go for, then obviously user-centered design is going to help you get there. And I like human-centered design even more than user-centered design. So you can use those two terms interchangeably, but human- centered design is also often used in a broader context because it doesn't necessarily mean that you're just talking about an interface and a piece of technology. Your solution for a human-centered design problem could be anything at all. And also there's more people involved in something than just your end user. There's whoever's telling you what to do. There's your stakeholders. There's all your people you're collaborating with, all the people contributing and your users. And they're all humans. And it helps to think about them all as people with different needs and wants and motivations, all as part of the same holistic view. So that's why I like talking about human-centered design. And the other very important thing is that it's not just about usability. That's a very, very important thing that should happen towards the end of a process, but there's so much more that can happen at the start of a process or before anything else happens. Cool. So my goal for this project was to get more young people to engage with politics. And that's very nice. It sounds very nice. And everyone was, you know, that's a good idea. You know, you should do that. But it's a good start, but it's a really long way from actually getting anyone to commit any resources or support or contributions. So what I started thinking about doing was mapping out this political landscape, thinking about who was it and what they all wanted. So we've got the political parties and there's big ones and there's little ones. We've got activists and big groups and organisations who have issues that they care about. We've got political commentators who are interpreting everything that's going on. We've got the media that's collecting all this stuff up and being it out again. We've got universities and academia who are interested in this from a long term trend sort of perspective. We've got civic hackers and coders who want to do cool stuff. And then down here we've got the disengaged young people. And lots of them aren't interested, but even more of them don't even know that this is all happening. In a Massey University survey, a couple of months out from the election, 92% of the people they asked didn't know what the date the general election was. So, yeah, sometimes they're not interested, sometimes it's just completely off the radar. So the next thing to think about was that's everyone that's there. What is it that they actually want? So political parties, very easy, they want votes. And they want a really easy, convenient way of engaging because they're very busy at that time of year. But also these little guys, little parties, less media opportunities than the bigger parties. So they're willing to work a little bit harder for their votes. And that was something that was useful. Activists and organizations want attention for their cause, want to raise awareness and get as much leverage as possible. These political commentators were looking for these big meta narratives to put everything into. And I noticed that one of the things that people were talking a lot about was the way that social media was changing political communication. So it's like, okay, tie into that. Media wanted stories. Often lots of their stories were about how young people weren't voting, which is unhelpful actually because the research shows that the more that you talk about the fact that people aren't voting, the less likely people are to vote because we'll do something if we know that everyone else is doing it. But if we keep hearing that nobody's doing it, then we're less likely to do it. The universities have research themes. So Massey University, I noticed, had a research theme about 21st century citizenship. And the civic hackers were interested in ideas about transparency and accountability with government and with representatives and keeping things open. And so all of these different groups had something that I wanted from them. I knew that they could all help the project in some way. But now I'd sort of gone through all of us thinking, I could start to think about, okay, this is what they want. And the young people, probably the hardest ones to try and figure out what it was that they wanted because often it's just to be left alone, but I'll go into that a bit more. But my suggestion would be that it doesn't matter if it's a user or a stakeholder or a collaborator or a contributor, always try and make an offer when you're asking for something. And this, especially the first time, the first iteration of the project, we had no budget. We had no, well, I didn't have really anything to offer people, especially apart from, this would be cool. But it was great to be able to think that through. And I think people did get a lot out of it. And as an example of how that worked, second time round, by tying this project and with Massey's strategic research themes, we managed to get funding through them, through their strategic innovation fund. Because we made it clear how it tied into the themes and also how it was going to build their reputation for civic responsibility. So always worth thinking through that stuff. So that takes us back to what it is that the young people wanted. And how would we get them to actually use the site? So the first piece of research that I did was sat down with a bunch of 18 to 24 year olds and we made some timelines. And I just asked them, what were you doing, thinking and feeling in the lead up to the election last time round? Where were you getting your information from? Who were you talking to? And where were you losing interest? Where are you getting interested again? And this was really just a way to structure a conversation. And we found out it's a really useful, well I found out really useful things. One was that people will really get their information through Facebook in this age group, even if it was just news article links that they're finding them because their friends were sharing them on Facebook. So that was important. Also people were getting really annoyed at the media about this point here. And that was because it was so, lots of the coverage is so conflict-oriented, so petty. And people really weren't enjoying that. But also that some people who were saying, I'm not interested, it's not that relevant to me. But yeah, I'm still going to vote because you can't have to. You know, everyone does, it's important. And that just showed to me how important that social pressure is, even if you're not interested. Cool. So the next thing I did was put together some personas. And these were based on some of that research that I just done. And also on some qualitative research that the Electoral Commission commissioned in 2007. So we had The Evangelist Cat. And I used cats because I didn't really want to focus on demographics like age and ethnicity. I just wanted to think about people's behaviour and the attitudes. And also people like talking about cats. So The Evangelist Cat is always engaged in issues even outside of election time. They've probably got a specific thing that they're really into. And they're looking for tools to help further their agenda. The Upstanding Cat is the good citizen. They're going to vote because they know they should. They're going to do their duty. They probably don't really like getting in heated conversations about politics. But they'll read up around election time. They'll make sure that they're informed because they know that that's what they've got to do. And they're pretty confident about how to do it. This is The Uncertain Cat. It might be a first-time voter. They're not sure about the process. It's all a bit scary. They really don't want to look dumb because it's quite confusing. And if something came up that meant they didn't have to vote, then they'd be quite happy about it. But if somebody really encouraged them or made it easy, you know, they'd probably do it. They'd probably do it. And talking to young people, they said, oh yeah, that one. That's the one that could really use some help here. The uninterested cat is probably never going to vote anyway. It's nothing you can do about them. But if you can get The Uncertain Cat, that's probably where you can do the most good. Yes. So what I did with the personas was put them together based on some observations and then show them to users and use them as a discussion point. And I was saying, do you recognize these? Are you one of them? Do you know people who are the other ones? And also, who's missing from this list? And it was a great way for me to check my assumptions. People figured out things that were missing, like the apathetic evangelist, the non-voter who's very passionate about non-voting and will give you a big speech about it. So that was good. It was a good way for me to sort and store all of my research. So everything that I was noticing about users, I could sort of file it under one of these things mentally. And if something came up that didn't fit in, then that was, okay, maybe there's someone I've missed here. And we used them to make decisions later on. So, for example, we were deciding which content to prioritize on the homepage. And we had this trending algorithm. And we were like, okay, do people want the newest, freshest, like just ask question or do they want the all-time favorite most voted for questions? And we thought back to those uncertain cats. And one of the people I interviewed had said, yeah, they'd want to make sure that, you know, they'd kind of covered their bases and they knew everything that they should know about. They'd want the top 40, if politics was a top 40. So we thought, okay, we'll make sure that the stuff on the homepage has some weight to it. It's not just the newest thing that you know that you're looking at something that other people care about as well. And it meant also that we could stay in touch with our users. So me and the technical lead, John Lemon, could have these conversations about what the uncertain cat needed. And it was a great way for me to take this sort of vague research that I've been doing and put it in a form that we could have conversations across the team about what users wanted. And I feel like this would help my dad when he was struggling to understand why stakeholders were telling him to build all this stuff if he knew about the cats. So they're really useful. They're really worth doing. Cool. So how to make them a couple of points missing off this list. But if you're interested in making personas, there are some good things online. I saw that usability.gov, which is the US government's usability website, have a good guide. But what I did was basically just talk to lots of users, helps to do some structured interviews, because then you've got, you can start looking for patterns and differences across different groups. Real quotes really help because that way they're authentic. They feel authentic. If you have any analytics or feedback from an existing platform, then definitely plug that into it because it's great. You know, if it's you've got real data to support what you're what you're observing and prioritise them. So when you make personas, it doesn't have to be every single different use case. It's just the main important user groups. So figure out who it is that is important and then show them to people, show them to your target audience and see if they resonate with them. And then show them to the rest of the team and make sure that everybody knows what an uncertain cat is. So once I'd done this research and also a lot more kind of mind mapping and messy design thinking stuff, we got to the point where we had three very high level user needs and that really guided the development from there and all the features. So we're thinking I need information to come to me where I am. And that meant on Facebook wasn't enough to hope that people might go to this website. We needed to get this content onto Facebook and Twitter. I need information that's easy to understand. Everything's really confusing and there's a lot of jargon around elections and politics. And along with it easy to understand we knew that the site had to be really easy to use because the more effort that your users been trying to figure out how the buttons work the less brand space they have to figure out what the difference in policy platforms are between the three center parties. So we really had to minimize the load on them there. And also it needed to be engaging. So people needed to feel like they had a role to play. It was relevant. They could use a voice and also that they could see that everyone else was also doing that. So this my map here is those three goals. And then we started brainstorming features relating to those goals. And from that we got onto our wireframes. So here we had pretty clearly defined concepts. Question answer platform and we knew why we were doing what we were doing. Which was great. And I'm not going to go into the wireframing stuff much because it was mostly John's really beautiful interface design. But he spent a lot of time going through and just refining and refining this. So it was really smooth. But we started out using both Sarmic to do our wire friends. Which was good because it really forces you to be lo-fi. You're not going to start getting caught up in colors and gradients and all of those things. And John also did a great job of managing our backlog. So once we had all the user tasks as Post-it notes on the walls he went through and put them on Pivotal Tracker. And this was really helpful. We had about half, he did about half the development work and the other half was done by volunteers mostly through Code for New Zealand. So they were able to just go in and really quickly see what it was that they could do. And that made it easy for them to contribute. So then we got to what I would call the rep workflow. And reps are what we called candidates or representatives. And it's a lot harder finding politicians to do user research on than it is to find young people. So what I did here was just try and think through as a politician what is that process going to be like. And that's what they looked like. I was thinking, okay, there's the party secretary. We've got their party hat on. They go and they do this and they do this. And thinking about again what they were doing, thinking and feeling all the way through that journey. And what the moments were where they might need encouragement, or they might need to be reassured that this was a legitimate thing, or they might need to be reminded or prodded to contribute a bit more. And I think that the participation from the representatives was one of the biggest successes of Ask Away. We had 60 different politicians from nine different parties. There were a few days where they were answering more than 50 questions a day. A few individuals who just go on big sprees and just do tens and 30 questions at a time. And I could see the answers coming in at 2 a.m. in the morning. So it was really cool. And in the last week, we had all the party leaders, except for John Key and Winston Peters, all answering a question a day. So that was cool. And yeah, it was just good to see how open they were to that idea. And there were a few things that we did specifically to get these reps on board. So the first is called the low ball or the foot in the door technique. And this is a marketing trick that people do, which is basically that if you get someone to agree to a small ask first, they're more likely to agree to a big ask later on. So when we were asking the reps if they wanted to take part, we didn't say you must answer this many questions a week. And you must have this many reps from each party. We just said, would you like to make an account and answer a question? And after that, if you want to do more, we'll make it easy for you to do more. But the first initial ask was a small one. A unique contribution. So the best way to get the most contributions to online communities is to make people feel that their contributions are unique and valued. So when we were inviting these candidates, often it was that we got a few people before the platform opened. But once it was live, I'd get a question about transport. I'd email all the transport spokespeople and go, hey, there's a really popular question about transport. You're the person from your party to answer this. You're the only one that can do it. And that worked quite well. Have a specific ask. So if you can make it really clear what you're asking, people are more likely to say yes. I found that too when I was asking developers for help really early on, when I had a vague, it'd be cool if we got people talking to politicians. People were like, yeah, come back in a few months. But when I said, OK, it's going to do this and this and this, I was like, OK, that suddenly sounds manageable. And the most important one was FOMO, Fair of Missing Out. So we really leveraged as soon as one party had signed up, we were like, yeah, this party is doing it. And if anyone was not answering as many questions as we wanted, I'd email them and say, hey, I just noticed that, you know, you haven't answered many questions. Every other party has been answering lots of questions and it's really noticeable that you're not there. So that was a very useful strategy. A few communications principles. So again, figuring out what people's motivations are and then articulating value. So basically making it really obvious what they're going to get out of it. Use the right language. Humans really like things that are familiar. So if you pick a few buzzwords or like bits of terminology that people relate to, and it's going to be different for every different group of stakeholders and different groups of users and then use that language. People will like it. Use multiple channels. So there's an advertising heuristic that we have to see something three times before we even notice it. And so with getting the candidates on board, I went firstly, sent a letter to the party secretaries and then I tweeted some of the reps and then I asked anybody I knew who knew any of them to talk to them and then I went to any debates that were on and then went up afterwards and said, hey, I'm doing this website. You should be on it. And even though it felt like, sometimes I was being, oh, and I called them all a lot as well. So even if you feel like you're being overly persistent, it does work eventually. And yeah, appeal to the identity. So the first people that we got signed up were the people who already had a reputation for being good at Twitter or interested in engaging online. And so we went to them and said, you're the exact kind of person. You know, you're so good at Twitter, you'd be great on this. And that works. Make it easy. So it's the same as giving a really clear thing that you're asking for. And sometimes the best way to make it easy to understand is to use visuals. So this, for example, is part of how I convinced Matthew and my supervisors that it'll be a good idea to open source the project. They were like, we don't know what that is and it sounds scary. So I made them this picture that just explained that there's different sorts of things involved and open sourcing doesn't mean that anyone can see all of the user data, which they were worried about. And so yeah, you make a nice picture and everything seems a lot less scary. And the other way that we used that, visualizing things, was with our media partners. So we ended up embedding the feed onto Radio New Zealand's website and also onto their youth platform, The Wireless. And they paid us to do that, which was awesome because it boosted us and it also gave us more exposure and more leverage. And this is a picture that we took right to the first time that we made this proposal. And we didn't read any code, I don't think at this point. We were just like, look, it'll be like this and it'll work. And they went, oh, I get it. Cool. So yeah, showing people a picture of things makes it way more tangible. And then user testing. So obviously we did usability testing. Firstly just, we just had a PDF with a couple of interactive bits and we got people to click through it and make sure that they could use it. And then we also set up a test instance and got design students to ask recent graduates questions because we wanted to see if the community behavior happened as well. And really, really important thing that we found out from that was that the upvote button was not at all obvious to people. So a very easy thing for us to fix, but we easily missed it. So always, always, always do user testing. And then tested it once it was live as well. And what I saw when we just went down to the market with an iPad and showed it to lots of people. And I could see lots of people really liked it. Lots of positive feedback. But also I could see some people going, oh, ETS, carbon emissions, capital gains tax. And it was off-putting to them. There's all of this jargon. And we couldn't control that, but I could see that they were having this negative reaction. Just from all of that kind of language. And that was something that there were hints about in the analytics, but it was so obvious when I was watching someone looking at it. So just worth seeing. So results. We had 22,000 people visit the site in the six weeks. 100,000 page views, lots of questions. 16,000 votes cast across all the questions. Fix the different reps. 100,000 answers, nine parties. And at least one question a day. From, at least one answer a day. From all of the different parties, which was cool. And people said really nice things, which is always really nice. And we were on the news, which is really exciting. And also. That was great. Oh, great. That was great. It also turned out, went up by 3% across New Zealand. But of course, that's a good thing that happened. Lots of cool initiatives going on. So I'm not going to claim that one just to ask away. But it was good that it happened. So what's next? So a huge thank you to Internet New Zealand at this point. Because they've just funded us to develop this into a resource for intermediate-aged school children. So we can put it in classrooms as a civics education tool. And what I hope is that having that restricted age group will make it way more accessible for the kids using it so they don't get that capital gains tax reaction. So I'm really excited. There were some schools using it already in the last election and enjoying it. So I'm really looking forward to seeing how that goes this year. So thanks very much. And if you're interested in talking more about this, get in touch. John Lemon and Womerton, who's Ron Crawford, were the main developers. Code for New Zealand is Rowan's open-source civic hacking brigade. And they're always interested in civic projects. And ask away some GitHub open-source as well. Cool. Any questions? Is this working? Oh, yeah. What do you know about the demographics of the people who actually use the site? Was it actually youth? Or was it people in their 80s talking about capital gains? Yeah. So it was actually the biggest age group with 25 to 34. That was 30% of the users. And the next, after that, were 18 to 24-year-olds. That was 20%. So not ideal, not perfect, but still a lot better than the average tenority at a debate in a hall, I think. Yeah. Well, it was interesting having the partnership with Radio New Zealand. We weren't sure if that was going against the Amelot Youth thing, but we figured better to make a conversation where everyone can join in rather than just young people. But we'll be interesting to see how that changes it, having it in a just young people context for schools. Do you know whether any mainstream open-source projects have done UCD documents and published them publicly? I don't know. Yeah, I'm not sure. You spoke a little bit about some negative user reaction in relation to particular content. Is there a way that you think you can deal with that? Is it about posting like explanatory material on the website, or is it just that the user community is sometimes going to be exclusive? I think that it'll be really interesting to restrict the user group. I think that will help. But the other way of doing it that I'm thinking is like I'm five on Reddit, just a framing thing. So I think there's a way that we can make it. We tried to do it to some extent, but making it really explicit on the website. This is for really basic conversation. I think that will be a good thing to do. I'm not sure if you know about the IDO, the Human Center for Design, or something like that. That's a good one. Yeah, that's a great one. One question. What did you do about looking at the number of questions, the number of answers, the sort like greens were huge there? I'm imagining one of the issues that a really good tweet said in the New Zealand elections, one thing it showed us, is how little different groups of New Zealanders knew about other groups of New Zealanders. And I think we're siloed. And if I had a poll of all my friends, like the greens were one. So what did you do to maybe get past some of those biases? Or do you think that was an issue? It was a difficult thing. And it was also difficult because we were encouraging issues-based organizations as many diverse different ones as possible. But we were encouraging them to use and promote the site. So again, you're going to get a certain demographic of people. But all of the paid marketing, which I did on Facebook, was all aimed at. People from universities who like Corona and Glassons and Helen's Sons were like some of the big filters that you get to get at 18 to 24-year-olds in New Zealand. So in terms of marketing, as much as possible, we were doing it broadly. In my position, you engaged all of us for youth, bloody youth. Yeah, yeah. The other thing is, are you doing anything to try and promote this whole tool for oversight, like to use it in other jurisdictions? So like overseas? Not yet. But I'd love to see it used overseas. If anyone here from Australia wants to do it, yeah, yeah. I think we can talk about that. Did you collect any information to determine whether it was actually the politicians themselves at answering the questions versus the staffers? Do you actually care which one it was that was answering it? And do you think your users cared which one it was that was answering? That's a really interesting question. Because I wasn't sure when I first invited them, like, do I make it obvious? Do I say this has to be them? But what I kind of think about politics, because we also thought, should we just have it that it's the parties answering the questions and not individuals? Because again, a lot easier to get that buy-in. And I can tell you that I know that some of them were definitely the politicians, and I'm pretty certain that some of them were staffers. I don't feel like it makes that much difference, because if you put something out in public with your name on it, it doesn't matter who else wrote it. Like, you kind of have to stand by it. And that's sort of how political communications work. I mean, lots of speeches. Speeches aren't written by politicians. But it probably would be a good idea to maybe, I don't know, maybe have that a little bit transparent on the site. I was interested to see, I didn't specify whether I expected it to be staffers or the politicians themselves when I invited them. So I was interested to see who replied for themselves and who did it for staff members. What do you think would be the biggest issues in taking this to, say, Australia or any other demographic? Or what focus points would they need? Yeah, I'd say anywhere where you've got an MMP system or a multi-party political system, it's probably gonna work. It would be, I was thinking about the US and I think it would be quite difficult anywhere where you don't have as many voices in the conversation. Because one of the main things that this offers is a way to see lots of different viewpoints side by side. So that's obviously one thing. Other than that, I think it would work. I think it would probably work in a local body context as well. See the first prototype of this was for the Meryl Elections in Wellington and that worked. So I'd love to see it tried out in different places. Categories we haven't built yet and probably would be useful to be able to sort by categories. If it was anywhere where bigger than New Zealand, I think that would start to become really useful. Well, thank you very much for your talk. Good to everybody here. So thank you to make for a talk.