Become a Creator today!Start creating today - Share your story with the world!
Start for free
00:00:00
00:00:01
Exploring Creative Data Visualization with Nadieh Bremer
75 Plays22 minutes ago
In this episode, I sit down with Nadieh Bremer to talk about her new book, Chart, and the creative process behind her data visualization work. We discuss how she structures the book around a spectrum from straightforward charts to full-on data art, and why uncommon chart types can communicate nuance so effectively. Nadieh shares insights from her time in a makerspace, how physical-making influences her digital work, and why sketching after exploring the data is so important. We also talk about client work, design decisions, and how she thinks about success in her projects. It’s a fun, wide-ranging conversation that highlights why Nadieh remains one of the most thoughtful voices in the field.
Keywords: Nadieh Bremer, data visualization, creative dataviz, data art, uncommon charts, Sankey diagrams, data storytelling, interactive visualization, static visualization, design process, sketching, makerspace, creative charts, PolicyViz Podcast
Subscribe to the PolicyViz Podcast wherever you get your podcasts.
Become a patron of the PolicyViz Podcast for as little as a buck a month
Get Nadieh's latest book, Chart, her previous book Data Sketches, and check out her website, VisualCinnamon.com.
Follow me on Instagram, LinkedIn, Substack, Twitter, Website, YouTube
Email: jon@policyviz.com
Recommended
Transcript
Introduction and Podcast Updates
00:00:13
Speaker
Welcome back to the PolicyViz Podcast. I'm your host, John Schwabisch. I hope all is well. I hope you are enjoying episodes of the show. I've had a lot of great guests on this fall, and I'm looking forward to getting a little bit of time off in late December.
00:00:28
Speaker
Take a little break, rest and recharge, but we're not there
Introduction to Nadi Bremer and Her Work
00:00:31
Speaker
yet. We've got another great episode for you to listen to this week on the show. I am joined by the one and only Nadi Brember. Her new book, Chart Art, if you haven't seen it, came out, um I don't know, a couple months ago at this point.
00:00:45
Speaker
Beautiful new book. focusing on Nadi's process, primarily creating sort of those bespoke non-standard uncommon charts. We talk about all aspects of her work. We talked about her process of creating the book, of writing it, of making all the visualizations that are needed to bring you a book like this so that you can learn or you can develop your own skills in the data
Exploring Non-Standard Charts
00:01:11
Speaker
visualization field. So if you are coming to this podcast,
00:01:15
Speaker
focusing on the standard charts that you have come to know and love line charts pie charts area charts all those things and now you want to branch out you want to try something different you want to try something that's never been created before you want to try new forms new colors new layouts Nadi and her book Chart Art and of course her previous book with Shirley Wu, Data Sketches are the books and the podcasts you need to listen to. It's really going to help you think about how to improve your own skill and your own practice.
Engagement and Past Interactions
00:01:47
Speaker
Now, before I get you over to my conversation with Nadi, I would be remiss if I didn't ask you to please rate or review the show. on iTunes or Spotify, wherever you get your podcasts. I really do appreciate those ratings reviews. It does help me attract more and more guests because people say, hey, this podcast has a whole bunch of five stars on it. I can reach my audiences by getting on the show. So I hope you will take a few minutes just to rate or the review the show wherever you get it.
00:02:17
Speaker
Okay, with that little ad out of the way, let's get over to this week's episode of the Policy Biz Podcast with the one and only Nadi Bremer.
00:02:27
Speaker
Holy moly. It's Naughty Bremmer. What a treat.
00:02:34
Speaker
Thank you for having me. it Thank you for that intro. I mean, it's been, i guess, since before the pandemic when I last saw you, right? Is that right? I think so. Yes.
00:02:45
Speaker
Was it maybe Data Sketches that came out in 2021? Did we discuss the book? I mean, I know. We did. We discussed the book. That's true. Okay. So I did see
Design and Structure of 'Chart Art'
00:02:54
Speaker
then, but virtually. But in person, I probably haven't seen you since I was in Amsterdam in like 2019.
00:03:00
Speaker
Something. Or it was we also went to Potsdam, I think. Oh, right. For Info Plus? Yes. But don't when that was. But also around that time.
00:03:12
Speaker
Yes, that's true. That's around that time. I also got terribly lost trying to get to Potsdam. i think, yes, because i like I got on a bus and I got off too early and then I got on a train or something and like the train was like stuff was just, and I don't speak German or read. It was very bad. It was bad, but I got there and I survived. So anyway, that's in the past. So good to see you. Congrats on the new book.
00:03:39
Speaker
Thank you. Very exciting. Very big. Yes. It's another big book. It's again, smaller than data sketches though, but it's still, you know, big enough to put on your copy. That's true.
Creativity in Data Visualization
00:03:48
Speaker
that's true. Is this like when you, i mean, I guess because of the images, but when you, when you start conceiving of a book from the beginning, are you like, this is going to have to be like, you know, 15 inches tall?
00:04:01
Speaker
ah um Not necessarily, but I do want it to have like a certain presence, I guess. But then no numbers attached to it, but just like I want to be like a proper book. Yeah. hardco that Someone's going to have to adjust their bookshelf so that it can fit on there. preferably Or have it do that like 90 degrees and then laid so it stands out even more. Yeah, that's right. That's right. Yeah, it's the one that's horizontal versus all the other ones. Yeah, I gotcha. Okay. And then hard to get because you've got other stuff on top of it. And then, but yeah, it's it's worth it.
00:04:32
Speaker
um Well, it's a lovely book. I have a few themes that I picked out personally, but I think maybe the best way to start to give folks sort of a preview of the book, and then I'll get you like your your take on some other more specific things, but like you have sort of like a kind of specific structure to the the narration or the navigation of the book is sort of fairly specific. because So can you just like give folks like an overview of, I guess...
00:05:00
Speaker
I mean, they'll read the book, so we don't need to walk them through it but like your perspective of why you wanted to write it in this particular way. So in general, I wanted to write a book about being more creative with data visualization and the things that I, that work well in in my past that I like to use, like my personal tips, and maybe that will help for somebody else as well. but And when I started thinking about it I noticed that, well, being creative really depends on a lot of things. And so I started seeing it as a spectrum, like, well, you can be creative with the more straightforward business charts, but you can also be more creative if you're making something where that's meant for like a newspaper or even a museum, you know, all the way on the other end. So I saw it as a spectrum between sort of pure, well, not not saying pure, like straightforward database and all the way towards like data art. And I thought that would actually kind of help to to structure the book. So there are four parts. And it basically runs along this line of like adding a little bit of creativity because, you know, you don't want it to over, yeah like overwhelm you in a way. Yeah. Just a little knots towards making it, you know, the main thing that that's all you want to do So the spectrum from database to data art. And that's, that's how the book is is set up.
00:06:17
Speaker
Yeah. It's an interesting, i love how it gets you sort of this like, period or exclamation point at the end on data art, which is like, you don't just have to do line and bar charts. Here are all these other things that you, Nadi, have tried. And oh, here's like, take it the next mile into this very fun, almost liberating like data art thing. Yes, yes, I find, I mean, data art, it can also bring it back into database where in data art, it doesn't matter if, you know, if you don't have to explain the numbers or explain exactly what everything means, you can just try things, even try things that don't really work well to visualize data. But then sometimes I realize that I can take part of that and then tune it back down and put that back into my data visual organization work to make that a
Nadi's Learning Journey and Maker Lab Experience
00:07:06
Speaker
little bit more creative. So yeah, like you said, it data art can just be freeing as ah as a hobby, as something you like to do, but it can come back to your work as well.
00:07:14
Speaker
And you took, as I recall, you took like a Maker Lab course? Yeah. Yes, I did. i did. you know, it was it was COVID time. Yeah. it's like What do you do? Yeah. um you're So so yes, I i took 20 weeks. It's a 20 week, basically half a half a year. sort of Yeah. No work. And just focused on learning how to design electronics, but working with laser cutters, 3d printers, CNC machines, and just, just making things physical. It was, I guess I needed that after, a you know, like a decade of working purely digital, just learning how to make things in real life.
00:07:56
Speaker
Yeah. Okay. So I have a bunch of questions about the book itself, but let me dig on the makerspace stuff. So. Did you find that that experience? So I guess kind of two questions. So one, do you continue to do that sort of stuff? Like, do you do you play around both like for fun and maybe for work? And then also have you found that that experience has helped inform the digital data viz work for clients?
Challenges of Creating Physical Data Art
00:08:22
Speaker
I don't do it that much still. and it it That's actually a little bit twofold. One is like life continues on if you're back into work and and you have to find the clients to actually sell it. So I have been trying to sell like more physical data art installations. But honestly, I haven't been successful. So it's only been personal things like so small laser cutting things, like making your own earrings. It's like that kind of level of stuff that is still... that's making my own lights yeah That kind of stuff. I still haven't been able to sell it to as ah as a client project.
00:09:00
Speaker
I think honestly, it might come down to, yeah, I don't really have it in my portfolio. So I first have to do some proper projects in my own. Yeah. But the physical projects also take just, you know, an immense amount of time and you need like the materials and the right tools. Like I don't have access to the laser cutter and see machine machine and the whatnot anymore yeah lee as I had in those six months.
00:09:22
Speaker
But they they have definitely given me a perspective on the the different kinds of projects. what's the word again, the different kinds of things you can run into that can go wrong when you have working on physical things. Like it's not as easy to debug a, like an actual physical chip where you make one of the actual little feet isn't solar dwell to the plate. And right. Yeah. Once that's messed up. Yeah.
00:09:50
Speaker
yeah Yes, yes. So i have a I have a big appreciation for how much more difficult physical can actually yeah than than before. And I still want to i still want to make like physical data art, but um but it's like it still it still remains on my wish list.
00:10:05
Speaker
Yeah. I mean, i I don't know what it's like there, but at least here, like commercial real estate, like, you know, I think also office buildings are not as full as they were a few years ago. So like finding the places to say, like, we would like a data art in our lobby where we had, you know, a thousand people coming in every day. And now we have 200 people coming every day. I can imagine businesses being like, well, maybe it's a different game now, but Yes. Yes. And then when you start talking with clients, they end up like, if you start explaining what comes down to it and then they're like, oh, so maybe, maybe we do go for like the printed data art piece instead of like the 3D fabricated one. Yeah. Right. Like, oh yeah. Let's just do a poster. Yeah. Right. Yeah.
Creative Chart Forms and Selection
00:10:48
Speaker
Yeah. Cause there are ah ah several pictures in your book of your work in an, like on the, on the wall of an office building.
00:10:57
Speaker
Yes, I do have a few of those. yeah I'm really, really happy. those are like Data art is one of my favorite kinds of projects, especially if it gets to hang in real life somewhere. Yes. Right, right. It's really big.
00:11:10
Speaker
Yeah, really big. Yeah. Okay. so All right. So we have these four sections in the book. um The first one that really kind of yelled out to me, unsurprisingly, is the section on uncommon charts. um And I'm going to read...
00:11:27
Speaker
the sentence or the two sentences at the beginning of that chapter that like really hit me. So um you write in the, in the beginning of the session to the book, I urge you to see if a non-traditional chart can convey the, that underlying story more effectively and with more nuance and care. Furthermore, because these often overlooked charts are less common, they can add a unique and creative flair that piques a reader's interest.
00:11:49
Speaker
um Which I think is like the perfect encapsulation of this kind of tension or trade-off between your like line bar charts versus these more bespoke different non-standard charts. um So I was hoping you could talk a little bit about your process, either on your own working with clients of this like decision to go, am i going to do line bar charts? Am I going to do something naughty Bremmer?
00:12:14
Speaker
Yes. yeah I think it's always trying to keep in mind, like, what is what are you trying to convey? And can I convey this better with a different kind of chart or more interesting?
00:12:25
Speaker
So the examples that I gave in this in that chapter is, for example, a Sankey diagram to show flows because it is so good. Sankey diagrams are so good at showing flows that you would say, like, there's there's no way to capture that really in a in a bar chart or a line chart properly. But also another one is a tree map.
00:12:44
Speaker
where in essence I'm trying to show how all of the costs of a company were split down into their subcategories. and You could do that as a stacked stacked bar chart. But a tree map has the same idea of sort of showing parts of a whole. And it does that actually pretty well as long as you're not interested in exact values and you're just mostly interested in, oh, this one's really big. That one is a lot smaller and this one is the smallest. It's kind of more like keeping that in the back of your mind like, What are you actually trying to do? And and these these smaller things that that make make a difference, like how exactly should people be able to read it and how quickly should they be able to read it? So always it it comes down to taking the sort of the situation and the audience in in mind and then knowing which sort of uncommon charts to use. Well, that comes down to sort of this mini chapter that I have as well, where it's about, you know, learn about the different charts forms that are out there. You don't have to reinvent the wheel, know about
00:13:42
Speaker
the Sankey diagram, network diagrams, tree maps, Voronoi tree maps. Um, and, and, Dumbbell charts and they're so like there's there's websites that have more than a hundred different kinds of charts. And if you have that in the back of your mind and you have a, you know, a base level of understanding when to apply these different charts, but you know, honestly just open one of these websites and just look through them to see like, oh, will this work, will that work?
00:14:06
Speaker
um And then knowing sort of, okay, so the tree map actually works because I'm showing parts of a whole and I'm showing a hundred percent. So that might be interesting. Oh, and I actually don't need to show the, you know, 1% after the comma, so I don't need a bar chart maybe. and then then this stream map could work really well. So it's keeping all of those those things in mind. That's how I kind of approach it.
00:14:27
Speaker
Yeah. Now the tools, the data viz tools have gotten, i think a lot of them gotten so much better that it's in a lot of cases easier to make some of these kind of non-standard on uncommon charts, right? Like raw is a good example. Flourish is a good example. Even
Educating and Communicating with Complex Charts
00:14:43
Speaker
some of the new stuff in Tableau where like you have these add-ins or whatever they call them plugins, stuff like that.
00:14:49
Speaker
For someone who's never experimented with these sort of uncommon charts, would your recommendation to be, if they're working those sorts of tools, to just like click and try and see if it makes sense to them?
00:15:02
Speaker
Yes, I would actually, if you have a little bit of time. i I remember when I first really started out in in data visualization, I didn't quite have a grasp of this data set could work well as this and that. yeah And I took all of these sort of basic examples. At the time, it was just you know of these D3 gallery examples of like a radial heat map and a hierarchical edge bundling thing that I never actually used in real life, but I wanted to try it. And I just plugged in my data and I just looked at it like, does it make sense? No. Does this make sense? No. Does that make sense? No. But actually, maybe the next time it will make sense for this new data set that I have. So really, it's all about trying and building up that experience that you have. But even if you are trying it, always come back down to, is this new chart form displaying the insights that I want to see? As long as you keep that in mind, I'm all for experimenting with any and all chartforms that you have. Even it doesn't work now, it might work the next time. And then you kind of have internalized that.
00:15:58
Speaker
Yeah. And how do you think about the, I guess, the, the, the challenge of educating the reader on how to read these like uncommon charts? I mean, you, you spent some time talking about it in the book, but like, especially for someone who's never tried a hierarchical dendrogram network diagram thing, you know, like what would you say to someone who's like, Oh, I actually tried this Sankey diagram. I've never built one before and it works.
00:16:27
Speaker
But my boss manager colleague was never seen one either. So how do I like teach them how to read this chart? That one is always like on a case by case basis. Sometimes the charts I find like the, the most used on common chart forms I find are kind of intuitive, almost like the Sankey diagram. It's just these. big lines flowing into different lines. It's kind of, there's something very natural about them that makes it quite easy for people to grasp. So as long as you put labels, like if you are teaching people how to read a chart, I would say text is almost your best friend, like annotations, labels, showing examples, making maybe like you have the main chart, but you have like three or four, like
00:17:16
Speaker
mini charts below that that highlight a certain thing from that chart so that you show like, oh, look at this specific line through Sankey diagram. It shows yada yada. And then you explain that and then you have both used that little little mini chart to show something interesting about the data, but have also used that as a way to explain people how to interpret this visualization.
Strategies for Presenting Complex Data
00:17:37
Speaker
So I like that as well. So text is really, in that case, your best friend and looking at hierarchies like proper, like the title and making sure that the title says something about what they need to understand, have a subtitle that explains like what is the data that you're looking at.
00:17:53
Speaker
um And then like with other uncommon chartforms, this also comes down a little bit to self-reflection. Like if you look at this uncommon chartform, was it with your particular dataset kind of easy to understand or was it very hard? If it was very hard, maybe this isn't the right case for it. or And then of course there's always the, it depends, but if it is it's hard to understand, but it conveys the data so much better than any other chart could because other charts, they sort of aggregate away the details, the complexities, the nuances. then maybe you should go for this more uncommon chart, form even if it if it takes a little bit of time to get used to, but then maybe do it in a different way. Instead of having this, only this one chart, the complex chart, start with some
00:18:44
Speaker
like easier charts, like little charts at the top that just slowly introduces people to this is what the data is that we're talking about. This is just some big aggregate numbers, some other things. And then you have the big chart or maybe on a separate tab, you have the big charts like, and now I want to show you the full complexities that you can find within this data set. And this is how you can see it.
00:19:04
Speaker
But again, text will there be your text regions really are sort of the the best ways or interaction if you have that. In sort of like Western languages, right, that are horizontal, that go left to right. Do you do you kind of assume that people are going to start kind of in the top left and work their way through? Because you just said like, you can have these little charts, say at the top before you get to the big chart. And that...
00:19:31
Speaker
So I guess my question is, do you assume that people are going to start kind of at the top and work their way in? Or are you using visual cues to say, hey, look at these, you know, whatever, three small charts at the top before you get into the big thing in the middle?
Interactive vs. Static Visualizations
00:19:47
Speaker
Um, yes, I do assume that but that also comes from I was thinking of more like maybe you have like a website. So you you see the little charts first, gotcha up to the big chart. And then yeah, the other way around first big chart and then the mini charts to show people how to like little nuggets of extra information that you want to point out from the big chart that I would put them at the bottom. But if you have more of ah a static thing or a one page thing, I think I would still put it at the top if we're talking about gentle introduction to the big one versus let's put them at the bottom when it's like highlighting parts of the big one.
00:20:24
Speaker
Gotcha. But i yes, i I do admit that that is a west a very Western point of looking at it. But I think like top to bottom, would say is pretty common any anywhere.
00:20:38
Speaker
Well, yeah, I mean, I think online for so for sure, right? I mean, I think the natural scrolling is vertical. I mean, I would, right, let's same with my perspective, it's still like English. i would But I think that sort of like just the way we use the internet, the way we use cell phones is still veric primarily vertical. Yeah. um Yeah, I hadn't thought about sort of the...
00:20:59
Speaker
the virtual, the digital version, you as the creator can kind of control in a way how someone's going to read the piece as opposed to the static world where you don't have as much control.
00:21:12
Speaker
Right. Yes. That always makes it a little bit more complex. so So let me ask this then. I've always wondered about this. So you have in the book, you talk obviously a lot about your your your interactive pieces, but then you have like a variety of examples showing you know your static pieces. And there's a couple of pictures of like a big poster on a wall with someone standing in front of it. So like, um do you...
00:21:38
Speaker
Technology aside, the creation, the digital coding, all that aside, do you find one harder to produce than the other because of how people interact with it and sort of where they navigate first?
00:21:51
Speaker
Yes, I do definitely find interactive to be harder. And I know that's, but that's really from um all of the things that most people don't see. And that's um browser bugs. It it comes down to the coding performance, making sure it works on a big screen and a small like mobile screen. That just explodes the amount of time that I have to do it. um And whereas I think thinking about how to lay out the piece, of static versus interactive.
00:22:22
Speaker
Like the static, you you have ah a bit of a a downside over interactive, I think, because I mean, you can't work with the scrolling, you can't work with animations, scrolling tellers, those are just you know gold in terms of slowly introducing a story, making sure that people only see what you want them to see. Where static, you just have, boom, that's it.
00:22:42
Speaker
But even the difference in complexity, the just pure in making it interactive on the web, as is nowhere near the complexity that making a static visual as where you try and make that as much as like following a certain hierarchy that your eyes will follow. Because those things are a little bit more structured, like big texts or some, again, in the Western world, something in the top left draws the eyes a little bit first, except if there's just this giant vibrant thing in the middle, then maybe they'll look there first, but then their eyes will still quickly go to the top left. It's just, you know,
00:23:18
Speaker
see if there's like a title there. So there are things there that you can you can just apply a little bit more stricter rules. ah not Not strict rules, but like general, how you say that, guardrails that you can try and use if you want to lay out your static piece. So, yeah. When you work with clients, is it...
00:23:39
Speaker
um I assume this is generally the case where they'll come to you and say, we want something for our website um or we want a, you know, a static document or report or thing for our wall. Does, does that ever like you're like, well, they wanted this interactive, you know, you'll be working and creating this interactive thing, but actually think like after some discussion, like a static thing would actually work better.
00:24:04
Speaker
Or is it generally like they want something and like, they sort of are you know generally like that's what you're going to end up building for them.
00:24:14
Speaker
No, there's definitely more than enough cases where I would advise a client to go for something different. And sometimes it's like you say, well, they have a certain goal that they have in mind and then they tell me what they want to and then I see the data and then I think, well, actually, maybe you should do this as an interactive instead because it's quite complex. We need to build it out in many steps or instead they say they want interactive because that's the cool kid in town and I say well maybe actually this is better static but part of that can also sometimes simply be budget if they have a certain idea of grandeur and then their budget is not quite quite there and like, well, maybe we start our simpler. And in other places, I just give them both options. I'm like, well, if I were to do static, I would do it like this. And if I were to do interactive, I would approach it more like this. And then I would give my advice. I'd say, well, I think this or this one is better. But then I just still leave the choice up to the client to make the final decision what they want.
00:25:18
Speaker
Do you have clients who you build the interactive piece, but then they want a static version and they sort of say, well, can we just take a screenshot of it and like kind of be done? and then you have to say, well, that doesn't really work. We have to do X, y and Z. Yes. Yeah. Okay. Yes, I also have that, yes. And then there are cases where i have sort of shoved in sort of a export to PNG function into the tool that also you know makes it into the right right dimensions because the you know the resolution on the web is much lower than when if you were to print it, especially if you print it like in a giant poster.
00:25:54
Speaker
Yeah. like There's complexities there, but also like you say, yeah, then I need to go, you know, add titles, add legends in different places. Right. There's, yes, I have had clients that, but, but I will say that most of them, if I explain it to them, understand that this is not just something that I can just plop on and it stays within the same budget. It's like, this isn't, and this is an extra scope. Like this is a scope and management and then will it will cost some money. And if they're okay with that, then, you know, I will do it for them. but right Right. Yes. Yeah.
00:26:24
Speaker
Yeah. Okay. yeah I mean, yeah it's a totally it's a totally different thing. Yeah. now But I don't blame him because like for for me, it's so obvious that interactive and static are two different beasts. But I yeah understand that a client who is not into database doesn't understand the complexities that lie within that. And that it isn't the same.
00:26:42
Speaker
Yeah. Yeah,
Amplified Encoding and Breathing Earth Project
00:26:43
Speaker
for sure. um One of the other... things that you talk about in the book is, well, you have this phrase amplified encoding, which I love because I think most people sort of use the phrase dual encoding or double encoding. And you make the explicit point in the book that it's like dual or double encoding just suggests that there are just two things that you can use and you're like, no, there's like way more than that. um Which I think is amazing. And I'm already like putting that into my teaching slides because i just think it's really good. um
00:27:16
Speaker
But can you talk a a little bit about like, I guess, like, you know, how you or think about or define amplified encoding and then maybe ah an example project where you sort of used more and more variables.
00:27:31
Speaker
Yes. So there is, like there's amplified encoding and then there is sort of adding more variables. But amplified encoding means, what I mean with that is that you take the same value, data values, which like just one one column, and then you apply it to as many different sort of visual channels that you want. Instead of saying like I can only use...
00:27:54
Speaker
this value to set the color of all of my circles. I also set it to the size of the circles and also set it to the opacity of the circles. And, and you know, and this can continue on as as as far as you want. Which I think because I, on the one hand, I think one of the the reasons for it is that by sort of stacking these different visual channels on top of each other, you you kind of give people more and more ways of seeing that value visually. So it's not just a circle that is big, it's a circle that is this big, fully opaque and bright red versus a circle that is small and very dim and you know grayish.
00:28:38
Speaker
That just stands out much more than if it was just the size or just the color. or just transparency. So that's one of the things I really like it. But another reason why I really like it is because it touches upon something that I mentioned several times in the book, and that's it increases visual diversity. So it makes this thing that you've created a little more, it has more for the eyes to look at and be intrigued by, especially if you want to make it a little bit more creative, it really helps to have more diversity, more not just circles that are only different in size, but imagine them also being a different shades of green from yellow to green and and like stand more. And so that's the the project that i have used this on, I think the most or like it's the entire project is one full amplified encoding going on. And that is breathing earth.
00:29:29
Speaker
which is a project I did as a personal project actually. And it basically shows a map of the earth with about 40,000 circles on the places where there's land. So it's like a grid of circles and only the places with land have an actual circle. And they are they are encoded to show the amount of greenness as measured from space.
00:29:51
Speaker
And so like I said, all of these circles and the more greener that that place is, the bigger the circle, the darker green the circle and the more like um opaque the circle is. So you can definitely, and then then there's an animation throughout the year. So it's like, You see the northern hemisphere like like blossoming up in dark greenness, getting getting bigger towards the summer and then then going back down again towards the winter. It just gives this much more like ah vibrant, likes it draws you in much more than if it just had been black circles getting bigger and smaller. or Right.
00:30:28
Speaker
just green circles getting, you know yeah, it's yeah, that's so that's why i think that's such an ah like, and there are so many visual channels. And I sometimes find with these charts that people are not employing all of each, like, they don't have to use all the channels, but they can use a little bit more than what they've been using. And I would only have, I think, a positive effect in most cases.
00:30:50
Speaker
Yeah. And do you find that you, do you find that you yourself need to like temper yourself down a little bit? Like you had like all these encodings like, whoa, like that's, I've got now color and opacity and I've got like, I've got patterns in it and I've got side, like, you do like, is your process a little bit of like, I'm to throw everything into this this and then I'll pull back a little bit.
00:31:16
Speaker
It, I am more of the latter. However, however, I would say that I am looking at me personally, I like to use a different lesson that is add more variables. And that means that I would say, well, there's greenness, but there was also what other variable, like the amount of like and the amount of wind that happened at right yeah yeah like again the The temperature at that point in time. like i I always like to put in more data. um And that means that I'm also using these visual channels, so I can't use them to do this this this amplifying encoding anymore. So it's always a battle between these two, but I am i am little leaning a little bit more towards more variables, but I i am of the person that just tries to put in as much as possible and and at some point feel like, oh, this is getting a bit too difficult or too overwhelming. And then and then i then I pull back and say, well, maybe this one isn't needed.
00:32:08
Speaker
right Or this one, I'll make this one really subtle, but just so right like a little much. I don't actually explain it, but I know it's there. Yeah, a little little Easter egg just for you. yeah exactly.
The Role of Sketching in Data Visualization
00:32:20
Speaker
um The other thing I wanted to ask you about was sketching. Because I think a lot of people in the field say, yeah, like you should draw and you should sketch. And it can be paper on pen. It could be...
00:32:34
Speaker
an app, whatever it is, but like do some sketching. But you write about in length in the book. And I think in a very helpful way that like you start by working with the data, understanding sort of the data, the structure of the data, the basic,
00:32:52
Speaker
you know, kind of ins and outs and then go to sketching a little bit. Can you talk a little bit more about that? Cause I think that's a challenge that a lot of people have of like, well, I can't just sketch cause then I'm just making stuff up and I've wasted all this time. And like, I go to the data and there's this weird outlier, but I didn't know that. And so it doesn't really work. So how do you like, yeah, like what's your, what's, what's your process between the sketching part and then the, the data kind of analysis part.
00:33:19
Speaker
Yeah, so well it it really touches upon what you just said is where if you start making ideas before having looked into the data, I think that on the one hand, you will miss things because you don't know the data well enough or what's in it. And on the other hand, you could come up with ideas that just don't work with your particular data set. Even in the the way that I do it, I miss things and then will come across things that I need to sort of retrace my steps and do it in a different way. But that's that's kind of why I always start with this analysis space. I just need to get this sort of card house up in my head on what are we talking about? How are things distributed?
00:33:56
Speaker
What stories can I find within this data or conclusions? What do you know what are interesting things to highlight? and And what should I watch out for? Like you say, outliers in this place.
00:34:07
Speaker
So i you know I do the i do the the basic things, at least like summaries and you know, mean standard deviations, just looking at lots of bar charts, seeing how things are distributed.
00:34:18
Speaker
And then I go into the sketching phase because then I have the tools that I need, which is on the one hand, always making sure that I know my goal, like what are we trying to do with this visualization or with this data?
00:34:28
Speaker
And then knowing the data, like what variables do I have? I kind of know how each of these variables is is distributed and what stories there are. And then I started thinking about, well, how can I actually visualize that? And I feel that with these two tools, it makes me much, much more able to come up with interesting and working visual forms that I can then actually make.
00:34:51
Speaker
And i mean this is this is about making them more visible. like This is more for the you know the important visuals. You don't need to do this for every bar chart you create, but for the ones that you have to convince management or it's like for marketing or a newspaper or anything like you just want to have yeah stretch your muscles a bit.
00:35:08
Speaker
um Then I would say sketching is always beneficial. Yeah. um So I want to go back for a minute because you just said something that just piqued my interest. When you are, say, working for a client, and and so I'm going to put this for your in your line of work, it's you working for a client, but for lots of other people, it's them working for their manager, their boss, whatever. right And you're like, oh, I think this uncommon, let's just take a Sankey, make it simple. I think this Sankey diagram will be better than these three pie charts that are linked together, right? Right.
00:35:40
Speaker
To make that case to the client, do you sort of make both of them and say, hey, look at how this Sankey diagram does a better job? Even though it's a less common form, it does a better job of showing the data. Because I'm thinking about the data analyst who has the same constraint where they're like, oh, my boss always wants a pie chart with a hundred slices and I can make this other chart, right?
00:36:03
Speaker
That is better. But how do I make the case that this chart form that is less common, maybe less familiar, but is at the end of the day going to be more effective?
00:36:16
Speaker
Yes. I think they're the, just show them is the thing that works for me the best. And so usually I can get away with sketching. Um, so it's the idea for them because sometimes making it is just, you know, it takes a really long time. So right you just don't have that. You can't do that. So I, yeah I sketch. the idea that I have in mind, like, let's do it as a Sankey. So I kind of look into the data, I see, oh, well, this is really huge numbers, so I should make one, like one line really big. And then the rest is kind of like, average, I know, we're talking about like 10 different streams and blah, blah, blah. So i kind of try and make it as, as you know, how do you say that? um
00:36:56
Speaker
specific to the data as possible without actually having to make it. And then I show them. And what I also like to do then is show between one and three other good case studies of psyche diagrams where I feel like it doesn't really it doesn't really take a lot to understand what's what you actually seeing here. But this this is how like an actual finished piece could look and look how how well it's actually doing for this data set. And I think it would work well for my data set. That usually works for me, like nine out of 10 cases that works for me. um Although I usually give them like three options because
00:37:29
Speaker
They, jerk you know, you that that's what I prefer even more. So the Sankey might be one and then like, you know, the pie chart might be another if I think it would work at all. But, and it's not necessarily about comparing the Sankey to the pie charts, but it's like just showing how well the Sankey itself would work.
00:37:47
Speaker
um But both are fine, but it's always the balance between how much time can you invest into convincing your manager, um, And I think the, I did have a case where i couldn't convince my client, like on a certain
Client Relations and Success in Data Visualization
00:38:04
Speaker
idea. And then I felt like, no, this, I'm pretty sure this is really, this is better. And then i ended up making it, and it was a, it was definitely sort of a risk because yeah they could still say like, no, this is not what we want. But in the end they like, oh, I forget, I forgot the the the in English words, but they're like,
00:38:25
Speaker
Okay, okay, you're right. It was bad. They wanted to use it. Yeah, I mean, sometimes you just say, you know, sometimes the expert kind of knows what they're talking about, right? Right, yeah right. i mean And there are also cases where it doesn't work out.
00:38:42
Speaker
I mean, yeah sometimes as the expert, you can be blinded by the curse of expertise where something that's so super obvious to you, just just can't explain it to the client. And yeah, just like, okay, well, that's that's too bad, I guess. This time we go the more standard route.
00:38:57
Speaker
Yeah. um So on this sort of track, um you talk in the book a lot about like what is effective, what is successful. And I'm curious from there's a lot of ways that we could talk about success. So I actually want to just make this about you when you are working on a project.
00:39:17
Speaker
From your position of the creator, how do you view success? I mean, obviously the client is happy and they write you a check that is successful. But I mean, just from like a personal perspective, are you like, I made the Sankey versus the pie chart and they really liked the Sankey and so therefore it's a success? Or are there just certain projects where you're just like, you know, just the output, like what is for you like intrinsically, like what what means success for you?
00:39:47
Speaker
Um, I think it can be multiple different things on the one hand, you're right. The client is happy, but some clients can just really show how happy they are in their communications and The ones that are just, you know, where you feel like you can you can see their like glowing, smiling faces through the email, um yeah those that really makes it like a definite success for me. Usually they also also kind of you know go the extra mile for them because they just, ah it it's like, oh, I really like you. That's always a good feeling. I mean, yeah, yeah, yeah. sure Yeah. yeah um So that's that's a success. But there can be other things where it's about how people interact with it. That's also a success where it's like you share it online and, you know, getting more clicks. the The likes is like every person. It's a nice feeling, but I like it even more when
00:40:47
Speaker
um When it's like, for example, I made this visualization once about all the fights in Dragon Ball Z. And then I got notified that it was shared in, I think, the Dragon Ball Reddit page. Yes. And people were just talking about what they found and what they saw. And so that was made it successful to me is like how, as a fellow fan, that yeah people are really digging into this and finding insights and things and stories that they they had never really realized about these characters in this anime.
00:41:17
Speaker
That's another part of success where I feel that that even if it's only like a very small niche of people, but that those people really, really enjoy just learning about what i what I'm trying to show them with this with this visualization.
00:41:32
Speaker
right um And the final one is that that's one that one still happens to me with a project I did about constellations and constellations for different cultures across the earth, like how, you know, Western culture, but what about Mayan culture, Korean constellations?
00:41:48
Speaker
um And I still get emails like several times a year about um about school teachers, about museums, about like astronomy departments that ask if they can use the visuals from this piece to you know teach people that there is more than just Western Constellations. so That's another.
00:42:09
Speaker
I guess it's it's related to the one with the Dragon Ball it's like people actually using, like you you get this sort of really tangible result of people using visuals that you made to to enrich in their lives in some way.
00:42:22
Speaker
Yeah, that's awesome. That's awesome. um Well, on that note, um ah let's see. Thanks for coming on the show. um People can get the book at wherever they get their books, I assume, right? Yes. Okay. um And if they want to reach out to you to have you teach for them or hire you, what's the best way to get in touch with you They can either go to my website, visualcinnamon.com or email me at info at visualcinnamon.com.
00:42:52
Speaker
Love it. And I will link to that show notes. If you are listening, want get touch with Nadi, you should do that because as you heard, she's great. Nadi, thanks
Conclusion and Resources
00:43:03
Speaker
for coming on the show.
00:43:03
Speaker
This was lot of fun. I really appreciate it. It was, it always is. Thank you for having me We got to do it more often than every like five years. Okay. All right. Thanks so much. Appreciate it.
00:43:15
Speaker
Thanks for tuning in, everybody. Hope you enjoyed that. Hope you'll check out Nani's book, Chart Art. There are links to all the things we talked about in the show notes to this episode. You can also check it out on my website, policyvis.com, along with other great tools and materials and resources to help improve how you communicate your data. So until next time, this has been the Policyvis Podcast. Thanks so much for listening.