Shirley Wu

So today we're talking to Shirley Wu. Shirley Wu is a freelance data-vis designer. I know you have your own special term for it that you made up. If you can give us a little background on what you do, then be great. Yeah, so the title I made for myself because I work for myself is independent. Oh, there's been evolutions. So the latest evolution is independent data visualization designer and engineer . I feel like this is more just me being annoyed at people being like, "What's your title?" Because I don't know what my title is. But essentially what that means is I work with a variety of different clients, usually in tech, and media, and journalism. And I help tell visual stories with their data. I love that. And I think it's such a new field that even if you have a shorter title, no one's going to know what it means anyway. Yeah. So I love that it's long and descriptive. I'm just like, "This is exactly what I do for you." It's just like a read-a-sentence. And I know you weren't always this job title. So if you give a little bit of background, like how you got started, and how you got into the skills, I think that'd be really helpful. Yeah, of course. So when I first started my career, I actually went to a big data company as a front-end engineer. And that's kind of where I came across D3. And as I started using D3 in my work, and then I started going to the local Bay Area D3 meetups, that's when I was like, "Oh, data visualization is a thing." And this is a thing that I actually really enjoy. And so for my second job, I went to a startup, and it was an enterprise security startup. And I was my official title there. So my first official title was, I think, front-end engineer, I think, or software engineer. And I think my second official title was member of technical stuff, which is like the part is like, "Bait." But what I did for them is they wanted essentially a visual tool to help their customers kind of figure out how to secure their data center. So instead of doing it by tables or anything, they wanted a visualization that was kind of like part of their products. And so that's what I went and helped them build for two and a half years, I think, essentially I was a specialized data visualization engineer, I guess, is probably the title now. And then I think after that, I wanted to go and try out things on my own, mainly because I realized like I really enjoyed data visualization, but not as much enterprise security. So then I wanted to kind of try freelancing to see what other sorts of companies were out there, what industries were starting to be in need of data visualization. And that's how I started freelancing. And then, you know, like people always ask you for titles. And I'm like, "I'm a freelancer." And I make data visualizations. And they're like, "What does that mean?" And I'm like, "Well, I design and I code." That's the title. A little bit. I'm curious when you were working at these companies, oftentimes you're a developer or you're a designer, were you designing the data visualizations or were you getting designs and then implementing them? Yeah. So that's a really great question. So I think subconsciously that's probably hindsight 2020. That's probably one of the reasons why I wanted to venture off on my own, which is even at my startup, which was quite like hierarchically, like vertically very fluid. And there wasn't that much structure. And it had that sort of like everybody does it kind of vibe. But even there, I was slotted as software engineer. And so, like fortunately, I worked with people that were very much willing to take feedback. So like the designer that I worked with, he was really great and took a lot of feedback. The product manager took a lot of feedback. But even then my role was software engineer. But I was like very much interested in kind of the whole process. I was like very much interested in like the customer feedback and like why, you know, the PM was thinking about certain features. I was very much interested in how those features were getting, you know, like translating into designs. And because I feel like by the time it gets to us as software engineers, there's a lot of decisions that's already been made. That's kind of hard to develop for without the original context. So I think it was because of that that I was like, and in my personal times, I like I had been doing all these data visualization projects. And for all of those, I get to do the whole process, right? Of like going out and collecting my own data and doing my own design and doing the code. So I think not being able to do be not being able to be involved in the whole process in my full time job, that was probably one of the things where it really drove me to be like, okay, I really want to try and go off on my own and see if I can do this whole process with like different clients. Yeah, I love that. And that really resonates with me with like, especially with data. You can't just like say, oh, these are the requirements. Let's make a design and then hand it off to engineering because once you actually implement it, it might not be at all what the designs look like. You just really have to know what the data look like before you do the design. And it's more of like an iterative process where you need those tight feedback loops. And that can be really hard if you're like sequest ered, even in a small company to this one part of the process. How are things going doing freelance where you kind of have all the control? It's funny. Well, actually, I actually wanted to say I really enjoyed that part in one of like the talks that you gave about kind of how you prototype and how you iterate because like that really is very much how very similar to how I work nowadays. And that's a really like kind of astute observation you made where I think my frustration was due to the fact that like, I couldn't do that iterative cycle within that within within that environment. And I think that's really like a I think that's really a product of 2014-2015 data visualization of I think it was so still so new in kind of the business area that like I think and I think the the conversation hadn't quite evolved to. Oh, we actually need to involve the data visualization person much earlier on in the conversation to make sure that like what actually gets like what gets designed is very reflective of the data itself. And I think that's a realization that we have now and hopefully I'm hoping that it's something that's now like much more implemented and much more commonplace in in companies. That's my hope. I don't know if it really is, but yeah, I wanted to kind of bounce off of that. But yeah, for freelance now, I really enjoy the whole being able to do the whole process. But I think at the very beginning, what I didn't realize was how much I didn't know. I just kind of like, you know, like I was a software engineer, very much trained as a software engineer. And I'm like, I want to take client projects. And it took me years to learn how much I didn't like how much I did not know in terms of the design aspect, like the information design. I think there was like a good year where I was like bumbling around just trying like different things and seeing what sticks from like a design perspective. And I knew I made a lot of design full pause like for data is I'm actually really, I'm actually really surprised that I don't get called out more often for like the bad like design decisions I've made. So I think then, you know, trying to kind of make up for that. And then realizing that I'm really bad at the data side, I don't really know how to analyze it and trying to figure out a process for that. So it's been a lot of kind of a lot of self-studying, like a lot of realizing what I was missing and self-studying. And I think there was like a period of time at the beginning of the freelancing where because I was trying to like, you know, when you're a freelancer, you have to be like, I'm the expert. Right. And so there was like a while where I was feeling major imposter syndrome because I was like, okay, I'm an expert in like this one subset that I'm like trying to convince you to hire me for. But actually you're like hiring me for this whole process. And I'm not an expert in these. So that's a, I don't know if that answered your original question about finishing. I don't even remember my original questions, but that was really interesting. And I have so many follow up questions, but I guess my main one is, do you think it would have been better for you to wait a little bit longer or like, what are some things you could have done when you were working at a company to kind of solid ify all of these areas so that you would feel more ready. I don't think that waiting any longer back then, because I went freelance in 2016. I don't think waiting any longer back then would have helped because I think again, it comes back to there wasn't as much of a culture of including data visualization like data visualization was an afterthought for a lot of companies back then. I don't know if that's still true now, but like, I don't think even with kind of the openness of the startup I was at, I don't think, I think it would have taken me a very long time to get to the point where they would have, you know, kind of let me take, let me be fully involved in the whole process. And so I feel like, I feel like going freelance, let me get a crash course on the whole process where like it got crammed into like the space of a year to a year and a half. Whereas I think if I had kept on staying at a company, no matter which company, because companies just move a little bit slower, right? Like unless you're like a really nimble, like tiny little team, a company will always move much slower than an individual. And so I don't think I will have been able to get that crash course and I think it would have taken me much longer to get that same education at a company. And I think even though I was doing the whole process with my personal projects, the very fact that back when I was full time, the very fact that I did not have to design the personal projects for any specific client, any specific stake holder, meant I was only like designing for myself. And you know, when you're designing for yourself, you're like, I'll do whatever I want. So it really was that having clients really taught me what I needed to learn. Yeah, definitely, especially since data visualization is about communication, right? You're communicating data. So having that feedback from someone is really nice because if it makes sense to you, you know, even working on it, even thinking about the data, of course, it makes sense to you. But that doesn't necessarily mean it's like built in user testing, which I think is really great. Like, I've done both client projects and personal projects and I find client projects almost easier because you have this built in feedback loop and you have these constraints to work in. I really like that. Yeah, definitely. One other thing I wanted to ask you about is when you have a new project you're working on, what is the process that you normally go through? Like, do you start coding right away or do you sketch a little bit? How do you approach new problems? Oh, I like this one. Okay, so so yeah, so kind of what I was mentioning earlier about how I've had to figure out a whole process, kind of, during my freelance time. I think I did a lot of that while working on a personal project called Data Sketches. It's a project with my really good friend, Nani Bremmer, who's an amazing data visualization. I don't know her title, actually. She's also independent. So data visualization, person, creator, designer. And it's a project where, you know, we had, we chose 12 topics and we each did a visualization project for that for like 24 projects. And I think the most interesting part was we required ourselves to do a write-up per project. And I think that really made me be able to introspect my process. And so from that, what I learned is kind of my process is very much like first get the data, then kind of like go through and look out what each of the metadata kind of is. And then from there, start formulating some hypotheses, some interesting questions, then take those questions and then kind of use, I personally use Vega Lite as like a very easy new charting tool. And then kind of like plug the data into Vega Lite into different charts to kind of see if those questions have good answers. And when they do, when I kind of find that central message, that's when I start actually designing and sketching my ideas. And once I sketch, then when I feel satisfied, I start coding it. And of course, like once I code it, sometimes it turns out to be good. And sometimes it doesn't. Then I have to go back through the whole process. And that's kind of like the iterative process that we've talked about before. And I guess I guess I just wanted to say that, you know, the whole process I described is in the data sketches project, and also in the new book, the data sketches book that we have coming out in like much more detail if you're interested in any of that. So yeah, I got some sneak peeks of a few chapters and you guys go into a ton of detail. I think I learned some new things. There's tons of really good tips and tricks as well as like the overall process. So I think it's a really good book. Would recommend it. And I'm excited to check it out when it does come out . Do you guys have a release date for it? Yeah. Oh, thank you. And thank you so much for all the feedback you gave us. You gave us so much great feedback. Okay, is supposed to come out early February. I believe the exact date is February 9th 2021. But please don't quote me on the date. I can't wait. Well, thanks so much for coming on. It was really interesting talking about job titles and your process and your background. I think people are going to get a lot out of it. Yeah, thank you so much for having me. This was really fun. And I hope we get to do it again. Maybe you can have like a second course and then what do you like? Yeah, or you can do a course and then I'll come on yours and make it straight off. Yeah. Thank you, Mule.