(upbeat music) - Jason Forrest is McKinsey and Company's COVID response center, director of interactives. He led the effort in McKinsey to create a series of interactive visualizations of a variety of data during the pandemic. The data was mostly economic and jobs related but it included things like a survey of mask usage and a really interesting investigation of the emotional impact of the experience as well. So, Jason, thank you so much for joining me. - Thanks so much for having me. It's great to see you again Bill. - Yeah, so I want to get started by actually sort of taking a step back, going back to March of 2020. At the time you were a data visualization specialist that was your title at McKinsey. But then of course the world turned upside down everything changed and that was, all about COVID all of a sudden for everybody, but for you really more than most, because McKinsey decided to launch the COVID response center. And so my first question for you is what was the vision for this and especially from a data visualization point of view? - Yeah thank you, great question. So, McKinsey is known as a kind of an industry leader in generating a lot of expertise into various sectors of the business world. But, there's a certain way that we syndicate that kind of information. And I think with COVID a lot of our partners and senior partners got together and thought, how can we do this differently? How can we syndicate our expertise in a way that's going to be more approachable? And what's interesting is before I was even attached to the project, they had a whole vision for data visualization. I think initially it was a little bit more straightforward but when I got introduced to the team, I have a wide variety of interests. And so I think they could see my enthusiasm and just kind of let me run with it from there. - That's great. Yeah, McKinsey is, I mean they pretty much invented thought leadership which is what that is right. Do original research, communicate it out to the world, makes perfect sense. The topic that we're here to talk about today is, the theme is visualizing complex and or controversial topics. And I wanted to talk to you about it because when the pandemic started, a lot of people sort of jumped a lot of visualization type people jumped into starting to visualize some of the epidemiological data. And there was a pretty quick and hurdy debate about that around the nuance and the complexity and potentially risk that you might face if you're, trying to visualize this data that's incredibly complex and fraught. And I really liked McKinsey's approach in particular because they decided at least in part especially at the very beginning, it seemed like maybe a decision was made, well let's not go there with a medical staff. Let's start by looking economic information, jobs information. And so I did want to ask, like was that intentional to sort of avoid the fraught nature of this type of data? Or was it more about just sticking to the wheelhouse that is Mackenzie's natural type of data that you tend to work with being economic et cetera? Or maybe it's something else entirely. - Well, I think I left off a part about explaining what the COVID response center is and it was effectively really an attempt to explore data and fraught leadership, as you said on saving lives and saving livelihoods. So those would kind of naturally kind of pivot over into kind of healthcare COVID data versus kind of economic impacts of the pandemic. So what's interesting, is it all started with a map like many of these things do. I think like most of people's kind of COVID work did, and it started with a very lackluster map to be frank that we I think many people in data visualization were like, Oh, well, I can map that, that's like the easiest thing to do. But with some great voices in our community, like Amanda McCullough who went through to this amazing who's in the public healthcare sector, did amazing article about understanding the nuance of COVID data in specific. I think it really caused a lot of us to take pause and think, how do we find the right partnership and expertise to deliver this ethically? Right, and it's been fascinating because I've gotten like really deep into this kind of COVID mapping data now. And there are certainly different schools of thought about what's the best practice here. But what I've been also really focused on is pivoting away from a kind of a retroactive view. Like here's like the total cumulative counts for COVID cases and fatalities, and to trying to focus more on like, here's what's happened latest. Here's where it's happening today. And what's interesting is that this even mindset of trying to create data visualization's storytelling that evolved as the pandemic it has run over time, has been really interesting because there's been we've actually been able to do similar ideas in, like let's say for example we have a market market cap valuation tool. Which we made in may, but we actually realized that we could breathe new life into it in September because you know, none of us have really thought that we were going to be talking about this six months later with thought it was going to be more of, a two or three month window. And as time goes on this idea of like how do you track the moment has actually become something that's been really important for us? And we filter that into everything we do. - Ah, that's really interesting. Yeah, now that I think about that, I wasn't aware of that as a particular focus, but now with that lens I'm thinking back about the things that I've seen that you do and it makes perfect sense. I mean, so there is a currency to, a lot of what you've covered and the way it's presented. It's funny you brought up the Amanda McCulloch article which was for those of you who don't know, Jason in addition to being at McKinsey is also the publications director for the data visualization society. I'm also on the board. So we know each other well and that article appeared in Nightingale that publication. And so you were heavily involved, I'm sure in that. The creation of this article and, how went out and I can't remember the stats but that was a pretty popular article. I think it was people, it was very timely. People took it very seriously. In fact, this is the second interview it's come up in recent interviews I've been doing for this for this series so very influential. - It was published I think right around the 13th or 14th of March. So that would have been the, basically the first weekend that we became obsessed with the subject of COVID, right. It, had, a release or kind of a it's swept across the globe kind of different. I think London had it maybe a week before us. I think the work of John Byrne Murdoch, it helped us see that there was a trend and that we could be on a curve. If we were somewhere here we could expect to be here a few days later. And what's fascinating is that Amanda's article came through and basically helped everybody understand that it's not data that you just pulling from some database but it's like a health care system. And that healthcare system is made up of people. Right, and, everything about the pandemic despite whatever different spins of the data we may think about. There is a lot of nuance to this data that we as data visualization practitioners just may not know. So one thing that's very interesting as I can talk a moment about is that, we have actually wonderful experts medical doctors that we collaborate with that Mackenzie. And it took us a little while to find the right partners. But when we did, it was amazing because they could say, "Oh, well you can do this right with data visualization." And on the flip side, we'd say, "Oh but why wouldn't you do it this way?" And I said, well, you know if you're looking at a case prevalence, for example it's going to show you more of a real time understanding of where the virus is in comparison to the populations. And so the reason why I bring this up and this might be getting into the weeds is that, it's that partnership with a subject matter experts. And it's not just like handed off, like here's a document like go make a map of it. But really that back and forth that cultural and collaborative exchange which I think has made like a big difference in some of the work that we've done. - Yeah, the example I always think of in that context is all of those charts that you would see, the financial times, it was among those doing it. Our world and data are, we're also doing great work those charts where it's like a line chart and it doesn't start from zero cases. It starts from the day we first hit a hundred cases or account like that. Which are only a medical expert of public health person could possibly let you know, that's actually how we think about these things. That sort of that tipping point that we start to worry about. I actually wanted to ask you about, so at the beginning of the project, you're focusing primarily on these economics and jobs. If I remember correctly, that was sort of the first series of visualizations you created. And this was big, important information, millions of jobs disappearing industries shutting down overnight. But I wanted to ask you, if you remember what were some of the most important or interesting things that happened from your perspective of the work you were doing at that very beginning of COVID and the work you were doing? - Well, we have a whole arm of the firm really focused on kind of economic outlooks and forecasts. McKinsey Global Institute MGI, for short. We were able to partner with some experts a good buddy of mine, who's an economist over there now. And we were, again, able to like they could tell us what was going on. And they're really involved in the data and infrastructure of how a whole economic structure works. Right, this is their, this is their passion. And what was again, interesting is they wanted to say, well let's look at vulnerable jobs. And they said, "well, we have all the data." And I was like, okay, well let's have all the data, which, what story is important to say first? So it was the act of actually dis-aggregating what the data was, what the story, each what we call module is kind of an individual page. What's the story of each module trying to tell you. And one of the things that we did is we took this vulnerable jobs and we split it into three sections. A geographic view so you can look at it, on a state and MSA level so Metro area, we looked at it as far as demographics, because, as again a lot of nuance about demographic data as it relates to population. So we want to get away from all of that and just focus on like, who are the types of people, their income levels, and also their educational levels. And then we had a third one that was just really focused on States, not whether the map at all on site but with a tree map and a and a circle pack to really show occupations and industries. And what's amazing, is that we found that our users and our audience actually come and looked at the different modules for different types of data that would then help them make different types of decisions. Right, It's fun. It's one of those things that it all makes sense afterwards. But at the time we were just trying to streamline at all. And what we've helped people do is to really parse this complex, heavily nuanced field of data and come up with something that was able they were able to make a much more human intervention. - Yeah. That's a perfect segue, actually, humanity, human to my next question. One of the most interesting things you did and there was a lot of it, So this is not to say there's not an army an incredible collection of interesting things but, I just love the emotion archive. That's what it was called in which there was an exploration of the emotional impacts on 120 or so people, who essentially shared their thoughts and feelings with you on what they were going through through the pandemic. And this is referred to in current parlance as data humanism. And it, it really did a great job of communicating the data, right?, The trends, the key emotions the regional differences, but also those really human stories behind what people were experiencing via video clips and quotes. And so, also just seeing what was the motivation behind that? Where were you trying to go with it? It's so different from the others. It's sort of a, taking another step in a really different direction for most of the rest of the collection. - Well, thank you for bringing it up because just as you said, we're focusing on lots of statistical data about jobs and U.S. unemployment. I'd love to talk a little bit more about that. We're talking about lots of very detailed County-level case prevalence and mortality ratios and all kinds of wild stuff. But we really felt like there was such a human story. I mean, everybody's spent a week or two staring out the window wondering where they were going to get their groceries or if their friends were safe or if their families would be safe throughout the pandemic. And so, the other thing that I think we're all kind of seeing is there have been the great change into how we work has also created opportunities for us to do things differently. So working from home is a clear example. I know that a lot of people had really had an unlock to their relationship with their families and their lifestyles by working from home or more freely. At the same time, just as we start to do research and McKinsey decided to do an ethnographic research study of cross 122 people in eight countries in 22 cities. And when I heard about it, I heard, I knew from the very beginning that there was going to be something different because it was much more kind of like open questions. Show me an object in your life which has come into to make, to be more significant in your post COVID life. Like something very open like that. Talk to us about your thoughts of your personal finances. Very different question to has your income increased or decreased, right? - Yeah. - And so we wanted to try to leverage this data humanism concept to visualize all the aspects of what this humanity could be. And to be honest with you, we had no idea what we were going to get into. Right, When we started, we were just kind of like digging around in the dark, trying to figure it out. And, we then ran into the idea of Robert Plutchik's a wheel of emotion that's basically connected to eight emotions that have three different variations. We were then given access to an ocean of video and diary entries. We went through a very rigorous procedure to get clearance for all of this, this content. And then only really tried to do is create something that was memorialized these people's lives and in the act of doing so we kind of thought that it would show more differences geographically and it didn't. It showed the similarities of the challenges that we've all faced in our individual lives. And one last point, and then turn it back. There was this one amazing, there was a moment that I had when I was digging through this giant archive of images and it was a woman in China, She was doing exercises on the floor and her baby daughter is draped over her shoulders. And in the background, you can see, all the kind of things stacked up. I think there's even some tissues on the corner. It's not like a, like an idealistic an ideal situation, but I thought there's so much data in this image. How can we, how can we just, how can we be respectful at presenting that data to the world? And I, it's really resounded with a lot of people so. - Yeah, I would love to see a little machine learning to explore the text of all these interviews, the transcripts of the videos, those images themselves. I didn't even thought about that, but yeah. I can't imagine the treasure trove of information that's in there. You sort of just touched on something and I did have a question around this, which is, what have you learned going through this entire process? Key lessons you can share on working on such a moving target. And it sounds like one of them is hey, focus that target on the current than now, like can actually sort of taken a point of view on it. And also, any other things that you've learned along the way of dealing with this complex controversial issue, like COVID. Cause that's, that's today's theme and in theory, and what sort of bubbled up out of that that allowed you to say, "Yeah the next time we hit something like this, "that's how we're going to address it. "Even very big picture terms is fine." - So, what's interesting about controversial data is in how you frame it and how you present it. I would say that we've not done anything controversial actually, because we've not really taken or positioned any of the data but really just tried to find an intuitive and transparent way to communicate it to people. For example, we just had an interactive launch yesterday that shows public policy interventions across all the States in the country at trying to slow the spread of the virus. Initially, people were nervous that it could be kind of misconstrued but what we did is we were able to basically find a way to show your sources, explain exactly what your definitions are, try to create an intuitive interface so that people can find the data and the information they're looking for and then lastly is you partner with the experts, right? You have them help you to understand what the most important things are. The other thing I would say that I've really taken away from this is what we learned so much in the emotion archive on data humanism, as that we've actually tried to apply that towards everything now. And what's fascinating is that, that doesn't mean that everything turns into a rainbow. It means that we have been very mindful about including more illustrations that would show exactly what a subject matter would be or we've even structured something in a way where people might kind of focusing on kind of a baseline user experience information design patterns, to try to create parallels where people can again find the information. - The other thing that we've done. And the last thing, sorry, is that, we've really tried to make each interactive novel. Right, and novelty is important, not because it's a novelty that's dismissed, but it it's, it catches your eye. It stands apart from the others because you're the one to figure out and understanding like that line between like where you hook somebody so they're really curious and they want to get in and explore the data versus just make something that's too complex that pushes them away, is. - Yeah - Is really the magic. And, I'm really proud of the progress we've made in this, in this front so far. - That's great, the phrase I always use is eye candy is important. We can make fun of it. Oh, it's just a pretty thing. It's no, eye candy matters. Cause it does, it attracts an audience, it is fresh, it is novel, keeps my interest, and yeah, if it doesn't make it too hard for me to use, then that's a good thing. Very briefly because we're running out of time. Have you been able to operationalize this within McKinsey? Like I know, you have this team working on this stuff and I imagine now you must be a well-oiled machine. Have you learned how to make this just run incredibly smoothly? - We've published 17 modules since April. And in that time, we've really tried to index more on like the design process. So we have basically a lead designer that goes through a whole collaborative session with subject matter experts that then prototypes that kind of proves that out. We syndicated with our leaders and then it goes into the development team all the way through even kind of QA committing code and then deployment. The concept of data visualization I think it's been kind of light on process. So, we've been trying to really kind of adapt concepts from kind of agile software development. So we have daily stand-ups. We have retrospectives every week so we can involve our process. We have a number of methods in communication, ways to communicate to each other to give each other feedback. To provide learner learning and mentorship opportunities to the various members of the team. So, it's also been really important for us to pull in more video content. So, initially we had these great collaborative sessions with these experts. But then, we realized that that their perspective was missing from actually communicating our work in the tool. So, now we're going to be going through and adding a series of kind of one minute long video tutorials. Where these experts let's you walk you through the data visualization itself. And it's fascinating. Again, I helped to build these things and I always learned from them as well. So, it's fascinating how we can learn and leverage new methods of communication in helping people to understand what we've made. - That's a great point. It makes me think of Hans Rosling when he did his TED Talks, he did his BBC video, the YouTube video of his bubble charts. And then that may, Oh, now I understand the story, I get it. Now I know how to use this thing. Okay, we're actually running out of time, Jason this is a great conversation. I really appreciate it. And I wanted, I know that you're you have a book coming out or you're working on a book about Isotype. And for those of you who don't know Isotype is when instead of like a bar chart, you have like a stack of icons and they're using a whole bunch of different ways too. But, so, I know that by following you on Twitter I get a preview of that book every day cause you're constantly sharing amazing good content on it. And when is that coming out? Do you have any idea? - Oh, well, I was about, I said that I was about three quarter of the way through it before the pandemic started. And then it's just slowed to a crawl because I'm juggling all this other stuff right now. I actually do. I would like to try to get it finished by the end of the year and maybe out next year. Just a little debrief Isotype as a way of creating representing statistics with pictures and icons. It was kind of developed in the '20s, really kind of widespread in the '50s and Europe in the U.S. and then it just disappeared. And my goal is to really get inside of all these different practitioners and understand the rules so that we can bring it back. And that's what the book is going to be about. The rules of Isotype. - Very cool, so we'll hope for a vaccine for a whole bunch of reasons. One of which is that maybe then we'll get a chance to see that book sooner than later. So, Jason, thank you again for coming here today. I really enjoyed talking to you about this and I do look forward to seeing that book when it comes out. - Thank you, Bill, I appreciate it.