Fiona (00:11)
Welcome back to Undubbed, I'm Fi
Sarah (00:15)
And I'm Sarah and we're continuing our Iron Viz special with even more of the Data Fam who've been brave enough to step into the ring.
Fiona (00:24)
Don't forget, it's best to watch this on YouTube rather than listen to it on Spotify or Apple Podcasts because you can then see the beauty of what people have designed as they step you through their process. We talk about design, analysis, and storytelling and see how the best of the best actually deliver for their Iron Viz submissions.
Sarah (00:48)
And of course, a few behind the scenes secrets that might just help you if you're planning your own submission.
Fiona (00:55)
or even doing some great visit work. So if you haven't already, hit follow and share this episode with someone who's dreaming of making that Iron Viz finalist list or maybe the best of in their own company.
Sarah (01:08)
and let's get back into it. This is Undubbed powered by DubDubData.
Sarah (01:13)
Welcome back everyone. Here on Undubbed, we're with Brian Moore, who has got an amazing Iron Viz submission that he's going to walk us through today. Brian is a four-time visionary, six-time ambassador, and leads the Boston user group. Welcome, Brian.
Brian (01:34)
And thank you for having me.
Fiona (01:35)
I'm so excited to have you here today while we were going through the submissions and trying to find them all on Tableau Public. Yours was definitely a standout. So thank you for coming on and joining us and happy Thanksgiving for the past weekend as well. I hope you had some great time celebrating and enjoying a break. Okay, let me get started. Your visualization is...
visually stunning. the dashboard is so beautiful and enticing to spend time in. But it's huge. It must have taken such a long time to pull together. Were there any things that help you to save time as you were going through your build? Do you want to share your screen as well? Awesome.
Brian (02:27)
Sure.
Yeah, so it definitely took a while to bring together. I spent a lot more time on this one than I think I ever have in past ARVs feeders. But there was one thing, it's probably not the most exciting of topics, but it did save me just a ton of time and a lot of aggravation. So I usually don't do much with backgrounds in Tableau. Like if you look at my Tableau public, it's usually just like, watch art.
Play it again.
But for Iron Viz, I wanted to experiment a little bit, bringing a background image. And if you've ever used background images in Tableau, normally you would build it out in PowerPoint or Figma. And you kind of have both things going at once, going back and forth. Every time you make a change in Tableau, you have to go back to that other design tool, make changes there, export it, bring it back into Tableau, make sure that all the sizing is right. And it can get really, really
It can take a lot of time, especially something like Ironbiz, where you're iterating a lot, and you're constantly adding more length to your dashboard, and then everything, you have to resize everything. It just gets really, really messy. To cut down on that, I ended up building the background directly in Tableau. Everything you see in the back here, like these dark blue sections, the lighter blue, these little colored dividers here, all of the lines,
are all built in a worksheet just using polygons.
So here's the sheet here. So this is the background for the entire dashboard. You can see, if you look really closely, those little divider sections. ⁓ So this is just set equal to the size of my dashboard. In the back there, I have a couple of parameters where I can set the size of my dashboard in case I need to change it. And then all of it's just controlled with a really simple Excel file.
the type of section I want, whether it's ⁓ one of those little dividers, whether it's full, meaning it's going to go across the width of the dashboard. If it's split, what percentage of the dashboard does that split happen? I throw in a height for it. And then pretty simple math to build out the rest of it in Tableau. And then so if you look at one of these sections, what
It's really nice about it. So say I have a section that I know is going to be 250 pixels, like this little section here. I know it's 250 pixels. I can add a section that's exactly 250 pixels. I know it's going to fit perfectly when I refresh that.
And then also, because I have everything in a vertical container on top of that background, I can set the specific sizes. Like you just see, you can fix the height of all these. So can make sure that it's exactly 250 pixels. So everything lines up kind of perfectly, just like it does in that sheet. And then say I have to go back and add 1,000 pixels to my dashboard. Normally, everything stretches out.
10%, you have to resize every single thing in your dashboard one by one. But because this is in a container and everything's fixed, it doesn't do that. It'll just add a thousand, know, blank pixels at the beginning that I can keep building. So it would save me a ton of time. And it probably, you know, only took less than an hour to build it out.
Fiona (06:06)
Let's.
Sarah (06:06)
Wow, that's amazing. Like I look at how much time I'm thinking I've spent like in Figma designing backgrounds and you're managing, like you said, two different tools and you're constantly going back and forth, looking at realigning the pixels. And the fact that you built that all out in an hour is incredible.
Fiona (06:26)
Yeah, I'm blown away as well. I love it's so simplistic in its nature, but I don't think that I've ever seen anyone approach the backgrounds like this. Recently, Sarah and I were working for an organization.
where we were replicating a lot of what was happening in Excel to start with. That was the starting point so that they could get rid of their Excels. And then they wanted to get into the visualization space. So think lots of tables. It was in the finance space as well. So, know, balance sheets, P &Ls. And they had ⁓ shading in the background that they really wanted to hit. ⁓ Probably not possible exactly with the type of ⁓ height
Fixation that you've got because sometimes they were removing the zeros But if they were willing to forego the zeros we could have perhaps got it exactly as they had it in there and their visualization rather than coming up with something that was more of a Compromise so we had to compromise and change where the shading was instead because there's only certain options that tableau actually give you
the one to five different options that they have. So I think this could have been another opportunity for us to look at it. definitely worth people downloading. Have you enabled people to download your workbook on Tableau Public? Yeah. So people could download the workbook and sort of take a look at it and see how it's all pulled together to use it because it's visually stunning the way that you've created these different areas.
in your story, know, in your long form. And I can really see quite easily both using the white space that you have here in that blue band, you know, bringing together the two columns, but also then the way that you've used the colors.
Sarah (08:24)
Yeah, I also know Brian, you've got another trick with colors. Are you willing to show us that?
Brian (08:32)
Yeah, definitely. Yeah, so because this was all built out in Tableau, it made it really easy to update or change the colors as I went. And I also just switched to the web browser real quick so I can zoom out a little bit. Down at the bottom here, there's this little choose your flavor section where you can choose between five different candy-inspired colored themes. So I made it really easy to do this by building up the background with Tableau
Click on the chocolate one.
Fiona (09:07)
Wow.
Brian (09:08)
Let's do a chocolate pattern, then we have cotton candy.
Sarah (09:14)
You know, my mind goes straight to imagine having this feature built in and you've got different clients that are on different, you know, different brand colors and you've got a template. This would be amazing to show them some kind of mock data and their colors really quickly.
Fiona (09:31)
I love.
Brian (09:31)
And so
I'm actually working on building out something now that doesn't rely on the Excel files. You can do everything in Tableau and build out something really similar to this.
people really rely on those other tools. I know some of those things are coming to Tableau, but I'm hoping to build some of that into a template. I think another good reason to do that too is if you're using templates at work and somebody builds a background image and then that thing just disappears, there's no way to go back and edit it. If you have to go and change that dashboard, then you have to rebuild it from scratch. It's really hard. ⁓
getting good centralization and collaboration with all of those different floating images or all those different background images floating around.
Fiona (10:22)
Yeah, 100%. And this coming back into the color palettes, I also think that maybe we do a lot of work with color palettes to try and make them accessible for people so that we're not having issues where we're excluding anyone from being able to read the dashboard accurately. But I wonder sometimes the sacrifice of going down the accessible route.
is that the colors aren't always as visually appealing and they balance. And so when you've got 94 % of the population or 95 % of the population that are seeing naturally the colors as they are versus the 5%, you know, who perhaps have that disability, maybe there's a balance as well. So you could have quite easily with that beautiful color switch.
have the accessibility view versus the standard view which may represent the client's color palette more ⁓ accurately.
Brian (11:26)
Yeah, absolutely. I haven't thought of that, but it's quite true.
Fiona (11:30)
So on your color palettes, do you use any specific tools to help you to find things? So you've got your chocolate one and your candy floss, was it?
Brian (11:42)
There's chocolate, cotton candy, candy heart.
Fiona (11:44)
Cod Kid.
Brian (11:47)
And just
Fiona (11:49)
And how do you come up with
the inspiration behind those palettes?
Brian (11:55)
⁓ It was all pretty manual, actually. I didn't find any really good palettes. If you just Google candy-inspired palettes, you get a lot of weird results. It's kind of like hand-picking colors from various palettes. liked fine-tuning them until they all came together. Yes.
Fiona (12:17)
Yeah, us.
Sarah (12:18)
Yeah, wow.
It's such an amazing visualization. We haven't got like deep into the data because I'm sure there's like a lot of time and effort that went into collecting your story as well, Brian.
Brian (12:30)
Yeah, so luckily, a lot of the research I had done already, otherwise there's no way I would have finished this in time, but I had done another viz on a similar topic like earlier this year. So I had a data set that got me most of the way. And then there was a lot of research that I kind of added to it, especially in this bottom section here, to get through like all the different acquisition chains to try to trace a candy brand from its creator.
up to its current owner. Some of them switched hands eight, nine times. It was pretty interesting. is a lot more complex than I anticipated when I started going through it.
Fiona (13:14)
Well there's so much that we could go into here like I can see the typography hierarchy that you use and really drawing things out in bold but we're out of time unfortunately today so perhaps people will leave in the show notes your visualizations so they can dig into it and we wish you well on the judging journey and hope to see you on stage in at TC next year.
Brian (13:40)
Thank you very much for having me.
Fiona (13:41)
Thanks for joining us.
Sarah (13:42)
Thanks,
Brian.
Fiona (13:46)
Varun Jain began his Tableau journey in 2021 during the rise of community projects in COVID. He quickly found his rhythm in real world fake data and rediscovering the fundamentals of great visualization. A former Tableau public ambassador, Varun's path has taken him from India to Dubai and now to France, where he's pursuing his masters while working as a data visualization engineer.
It's a true inspiration for anyone looking to shape their career through passion and persistence. And we're really excited to welcome him onto Undubbed today to share his Iron Viz feeder submission because it's absolutely stunning.
Sarah (14:26)
Welcome, Faroon.
Varun Jain (14:26)
Thank you so much.
Thank you, happy to be here.
Fiona (14:28)
Thank you.
Sarah (14:30)
really love the way that Tableau has taken you around the world, Varun. That's amazing.
Varun Jain (14:37)
Yeah, it's life-changing really. Never would have expected this to happen. yeah, mean, dream on folks. That's all I would say.
Sarah (14:45)
I love it. Okay, Varun. Now, quite often when we are building data visualizations and collecting data and looking at how to visualize them, Tableau isn't the only tool we use. And obviously, there's a lot of hype going on around AI as well. So could you talk us today through some of the tools that you've used to support Tableau along your journey?
Fiona (14:45)
Yeah, I love that.
Varun Jain (15:14)
Absolutely. So I think one of the biggest entry barriers to Iron Viz a lot of times is that you have to prepare your own data. You have to fetch it out. You have to figure out how to properly gather it from different sources for matters. it's a big headache. I'm from a technical background, but sometimes it's like it's too much even for me. So I can understand like somebody who's not like very good at programming or not from that kind of background will feel.
super discouraged to enter the competition. So this year, I actually tried to use the AI tools a lot in my visualization. So let me take you guys through it and you'll be, I think you'll be surprised of how extensively I've used this and how easily it helped me like go from an idea to a complete viz. So, okay, let me share my screen.
So here you can see my visualization. It's about the recipes that I've saved on Instagram. ⁓ I have a collection. I call it recipes. There's about 120 or something recipes in there. And I decided to of scrape that, get all the information from those videos and the captions and everything, and try to visualize it, try to see which ones I can make, how does it fit with my pantry, and things like that.
I think there's no easy way to do it. There's no easy way to look at the videos, write down all the ingredients, write down all the whole recipe and all the different metrics about it. So there's a few different ways I tried to approach it. There was, of course, a little bit of manual tasks. I have to, I will quickly show you my Excel spreadsheet. I have to extract the links. So there's about 100-something recipes. This one I did manually. I get all the links.
And then I went to my favorite tool, which is Clods, and I asked it to help me extract the different reels. I give them this data set. I ask it, give me, look at these recipes. Look at the caption. Look at what the video is about, how many likes, the different kind of metrics I can easily fit. And it really helped me with that. It's not perfect, but I have like
but it does go a long way to giving you something that you can build on. Second tool I used, which is very simple, similar is an agentic browser. It's by Perplexity. Maybe I can show you. This is how it looks like. It's called Comet. It's very new, it's very niche, but it's essentially, I asked this browser to watch those videos for me and actually write down ⁓ what's in the caption and things like that where
where Claude was not working. So I have like two different things in parallel going on and trying to get as much data points as possible, as quickly as possible. So this kind of reduced my extraction work by a mile. I don't think I can even put a number to it because otherwise I would have to watch each and every 120 different reels, maybe rewind it, fast forward it, or write down each and everything manually. this has been a godsend, honestly. ⁓
I'm pretty sure I would not have been able to participate in this year's Iron Viz if not for this. So data extraction is like super, super convenient with tools like these. So these are very easy to like take the data and just read it and do some manual things that you don't want to do.
Fiona (18:48)
Yeah, well, like, ⁓ so I find it incredibly interesting that the thing that we've started talking to you about is the data and the data journey, because obviously it's so important. It's what you're going to visualize. But the reason why it's interesting to me is obviously looking at the screen here and what we can see your visualization is so beautiful in terms of the design aesthetic that you have in here. But we all know that we spend
about 80 % of our time and data and getting the data right. You know, even in day to day life, let alone doing something like this. So it's really cool to hear you go into the process that you've used. Perplexity something that's sort of been in a distant radar to me. Do you use it often for other things?
Varun Jain (19:40)
Usually, Perplexity is like my go-to search engine for now. I almost stopped using Google altogether. I always ask for Perplexity, any kind of search query I want. If I want to go on a trip, I will just quickly ask, like, tell me what is good, but where can I go in the next two, three weeks or next month or something like that. It's like super easy. It gets you. It has its memory, so it knows what you're searching for. It's super convenient, and it just works.
Sarah (20:09)
Nice, We'll definitely have to give it a go. Varun, I see here, looking at your visualization, there's a lot of information here. And I really like the way you've restricted your color palette. Was that quite hard to do? Because you've got like, what is it, four or five colors and a couple of kind of shading colors there. That can be really difficult when you're talking about so many different data points.
Varun Jain (20:34)
Yeah, I think that's one of my guiding philosophies while building any kind of visualization. I try to have a background color, a couple of accent colors, and shades of it to use it in different ways. So here, my background color is this very light beige, brownish color. With my accent color is blue and orange. And then I just play with the shades, like dark brown for the text, and lighter brown for some bars, and things like that. It helps me like...
It helps me just have two colors, blue and here orange, to just help the eyes of the viewers to focus on certain points. Positive is blue, I'm sorry, and orange is yellow. So with just two colors, I have much less choices to make. It's a bit contradictory, but it makes my life easier when you have less colors than you have more.
Fiona (21:31)
It certainly does, but it takes a lot of effort to get to that level of restraint as well. One thing that I observe when I'm watching or I'm looking and sort of drawn into your visualization is the typography that you've used and also the hierarchy of typography. So I can see things like you're pulling numbers out and you're bolding them.
quite frequently and they're often larger than the rest of the text that they're adjacent to as well. Can you take us through what your thought process is in typography, especially also if you could touch on the different types of typography that you're using as well?
Varun Jain (22:20)
Yep, absolutely. again, like ⁓ as with color hierarchy, you have to be really careful about textual hierarchy as well. It does the same thing. If you're not careful about it, it distracts the audience. And if you're really thoughtful about it, guides the audience in a way that you want them to read through the visualization. So first thing that I usually do is try to force it on Figma. So it's another tool that we haven't talked about before, but it helps me go through the design process and just build the background.
Finally, and you can see I have some textiles here on my right. So I have specific one for chapter title, title, copy, then I have to emphasize something, so it's like the bold version. And I have quite a few here, but it goes from very big to very small. And I then don't use any specific manual fonts or manual sizes. I just reuse these specific styles, so it kind of forces me into this hierarchy.
So this is for generally how my backgrounds work, how my general copy and the text works. And ⁓ what are the principles behind it? I think it's very easy to find in the design world. You start with the big and bold for your titles. And as your importance goes down, the font goes down. And you also use color to highlight to certain points.
I want to highlight my chapter subtitles. I use colors to highlight my chapter titles. I use a little bit of these borders without using color. So it gives like a bit of a mix and match to play with. And also the sizes go down as you go down the hierarchy. That's also very important. So on the design side is this. I try to force myself with Figma. Now on Tableau side, it's really basics, I would say. Like you have your
bands very big and bold and you have your ⁓ labels very in the background. You don't want them to shout to the audience. You want the audience to look at them and then get drawn to the visualization first and then read the labels along with them. So it's easier on the eyes. ⁓ It's a bit harmonious as well and it just doesn't distract. It kind of seems into the
background and scenes with the rest of the visualization.
Sarah (24:47)
Really interesting and love that use of figma in there as well. I also see Varun that you've gone for a kind of, you can see everything on the page where we're seeing a lot of long form at the moment. So what was your choice decision behind that?
Varun Jain (25:05)
⁓ I just, don't think I do long forms code. that gives that, that would be my answer. If I show you my Figma file, I started with long form, like you see here, this was my first draft. I have a hidden version on my profile So I guess I realized pretty early on that I would not be able to do it. It's not my, it's not my strong suit. A lot of my other visualization are also.
horizontal, my previous Iron Viz submission again horizontal, it feels like a book and that feels more natural to me. So I guess that's the answer. It's personal preference, I would say.
Sarah (25:41)
Lean into what you do best.
Varun Jain (25:43)
Yeah, exactly.
Fiona (25:45)
And I was really interested as well to see that it read from left to right. I recall a similar submission a few years back from CJ Mays. Do you recall the visualization that he did? I think it was for Dari of his CEO.
Varun Jain (26:01)
I,
yeah, exactly. That is like my, that's my North Star, I would say. I had it in mind. I always have that within my mind. Like that was my inspiration. It's nice like you, catched up on it because I didn't think it was that straightforward. I'm sure I have it in my favorites. Anyway, it's very similar. It's like, it goes from left to right.
So I love this viz There's not enough good things I can say about this. It's very detailed and it's still very, very, very easy on the eyes. It has great analytics and I want, this is my inspiration, in less words, that's the only thing I would say. Like it goes from left to right. It goes through certain, a chapter format that I went through. It has some pop-ups.
Fiona (26:29)
Yeah
Varun Jain (26:52)
which I wanted to do, but I ran out of time. And I love how approachable this is. I think that's the word I would use it, and filled with detail at the same time. And I wanted to do the same with my business. Well, I think I reached a good enough level, maybe not as good as this. This is a beautiful piece of work. I'm a super fan of this. I'm glad you got it. I didn't want to.
Fiona (27:20)
I
definitely think that it's right in right out there with ⁓ CJ's version as well. And you've got some different points of difference as well. So just around the way that the colors are, I think that you lean into stronger colors to start with and then take trailing colors from that as well. Whereas CJ's taking more of a pastel view.
primarily in his visualizations as well. So it's really interesting to see that development. I personally, when you've got a background, I do like things to be a little bit stronger so that it's easier for people with vision difficulties to be able to see things as well. But CJ is an amazing designer and he has such a great design eye. I was really fortunate to work with him.
when we were at JLL, which is when he actually did the submission. So that's how I recall is, we were all really proud of CJ for making it to the finals.
So thinking about the finals, ⁓ Sarah and I have this conversation. Part of the reason why we don't even put into the feeder competition is it would be our very worst nightmare to be standing on stage in front of people competing. When you think of the Iron Vis stage or doing your submission, is that something that you would like to aim for is actually getting to that competition stage?
Varun Jain (28:35)
you
Fiona (28:46)
Or is it, I'd just like to make it into the top 10?
Varun Jain (28:50)
⁓ I try not to think about that because it's the same. You think about, you have to be on the stage. How can somebody do make a viz in 20 minutes? It's really hard to do it. It's really hard to think about it and try to do it. But yeah, I just try to put out my best work. If it gets into top three, top 10, great. If it's not, then I guess as I say, win or not, nobody loses, right?
Sarah (29:21)
Varun, I think you should be really proud of the work that you've put out there. It's definitely one of my favorites, although I feel I say that about a lot of visualizations. I wish you all the best. It's been great talking to you today.
Fiona (29:21)
I mean fish.
Varun Jain (29:34)
Thank so much. Thank you for having me. It's been fun.
Fiona (29:38)
Yeah, it's been absolutely fun. Thank you so much for joining us. And now we'll pass on to the next DataFam member with their Iron Viz submission. Thanks, Varun
Varun Jain (29:48)
Haha!
Sarah (29:52)
Next up, I'm delighted to introduce Kinsey Miller. Kinsey began her Tableau journey in 2019 and is still very new to the Data Fam. She's published four absolutely beautiful Tableau public vizes, which I think you should check out. They'll be in the show notes. And today we're going to look at her Iron Viz submission.
Fiona (30:15)
Welcome to the podcast, Kinsey. Yeah.
Kinsey (30:16)
Thank so much. Thank
you. so happy to be here.
Fiona (30:20)
Yeah, it's amazing to have you. in terms of our research for this podcast, we went through Tableau public and tried to find the visualizations that were tagged for the iron viz feeder. And I have to say, I hadn't heard of you in the Tableau community before. And so I feel like we've made an amazing discovery and I want to introduce you.
to our fun community and get you some publicity out there because your visualization is visually stunning. And it's amazing for me to hear that, you know, visualized data isn't your day job.
Kinsey (31:02)
Thank you so much, Fi I don't know, it's hard for me to take compliments like that. You're kind of overwhelming me right now, all the things you're saying right now, I really appreciate it. Thank you so much.
Fiona (31:13)
Yeah, well, we're to get you into the data fam where you'll get lots of praise like that and it will become second nature. So it doesn't feel so much like imposter syndrome because we really want you to be able to shine in it. Why don't we get started today? If you could share your screen and tell us about your favorite aspect of your Iron Viz submission.
Kinsey (31:39)
Okay, so I'm going to show you first the top so you can kind of get a lay of the land for the visualization. I have to say one of the things that I am most proud of with this viz is really the data engineering that went into it. When I heard that the Iron Viz theme this year was food and drink, that's really what inspired me to enter. As I was talking about earlier before we started the recording, Iron Viz wasn't really something that was on my roadmap for 2025.
It's something that has always just seemed so intense and I have a lot of respect for the people who enter it, which I learned is quite the journey. Getting from data selection all the way to the submission just a few weeks later, that's not something that I am very well versed in doing. All of the business that I have on my Tableau public website were months of work. ⁓ So to do this in a few weeks was something that alone I am proud of.
When it came to selecting my topic, I really wanted to do something that was nutrition focused. One of my very few visits that I have on my public portfolio is something that I published six years ago now, which was what I had called my coffee calculator. And the centerpiece of that had been a nutritional label and it was fueled by a number of really popular coffee drinks. I had pulled the data from, fastfoodnutrition.org was the name of the website.
in like the complete state for each beverage and the filtering and tableau would pull up each beverage separately. And I had always thought that it would be a really fun viz to revisit and kind of do my own makeover Monday on. And when the Iron Viz topic came out, that's immediately where my mind went. But I wanted to do something a little bit different. And one of the first things that came to mind was cocktails. Just because there are so many, it's a drink that you really can customize.
And part of that Makeover Monday vision that I had for the coffee calculator was how do I make this customizable? And so that is kind of where I started. Before I even opened Tableau, I was thinking through how I wanted to set up the data to make this possible in Tableau, to really enable customization, drink selection, and make sure everything was dynamic and still pretty quick in Tableau.
the way that I went about collecting the data was I started by researching cocktails. I'm not a cocktail connoisseur by any means. And I was really happy to find the International Bartenders Association, which puts out a list of cocktails that they kind of consider the industry standard for the time. And they have 102 cocktails that they identify.
sort of as the current Canon. So that was an amazing discovery because that kind of gave me this finite universe of cocktails that I could work within. So I started there and I created for those of you who are interested in kind of the data engineering side of this, I created an Excel file that was essentially one column was the cocktail name. The next column was the ingredient and then another column quantity. There were some other columns in there too, but those were the three that really mattered.
So went through, I collected all the recipes for the 102 cocktails. And then I took the list of ingredients and started a new file where I went through and I collected all of the nutritional information for each of the ingredients, which as you can imagine, a lot of data collection and research. And I will say I was getting kind of stressed part way through. I think it was not until week three of Iron Biz that I was actually opening up Tableau.
to get the data in. And I will say I was really relieved when I got the data in and I kind of, you know, reached the benefits of having done such meticulous data collection because it made the rest of the dashboard build a lot easier. I was able to kind of quickly build in the parameters that allow the user to change the strength of the drink. so this essentially manipulates the quantity in the drink for the alcoholic ingredients.
And then the sweetener does something very similar, but this is really catered towards the honeys and the syrups that go into the drinks. So as a very long-winded answer to your question, one of the things that I'm most proud of is the data and just the quantity and quality of the data that went into the dashboard. But from a visualization standpoint, one of the things that I'm most...
happy with in terms of how it turned out visually, but also in terms of the information that it conveys is what I can best describe as a heat map meets histogram. And with this, it is essentially a histogram, but instead of height of bars telling you how many drinks in this case are in each bin, it's the color saturation. It shows you how many drinks are collected in that bin. And so this was something that
I thought fit my Viz theme a lot better and was just kind of more succinct than doing a histogram because obviously I did want to do three. So alcohol, sugar, and fat. So I was really happy with how these turned out from the beginning, but then I had the idea to kind of include a pointer for the drink selection. So this was kind of a hacky tableau thing that I did, but I created a separate sheet.
that used the same bins as I had used for each of the histograms. And then I used some calculated fields that were driving the color. So if you look in the workbook, each of these triangle pointers actually sits within a set of triangles. So there's a triangle centered under each bin, but then I color coded them using a calculated field to just highlight the one that was directed at
the drink that the user had chosen. So I haven't seen anything like this in Tableau before, so I, it was kind of hacky getting there and I wasn't sure it was going to work out. But when I got it together, I just was really happy with the way that it kind of served my dashboard. while also I think being a really pretty element.
Sarah (38:00)
I would agree. It's a really lovely way to show, what alcohol, sugar and fat percentage that in this example, cosmopolitan is. And I really love in your design, how you kind of kept that nutrition facts down the left hand side is like a label we'd expect to see. And I think looking at your
What do we call that down the down the bottom right hand corner? It's like a flattened bar chart. I'm not sure. Like I really like because if you were using the heights, I think you'd really detract from the result there.
Kinsey (38:35)
I thought so too. was trying to think through like, if I wanted to do some sort of like parameter sheet swapping to kind of let the user choose which element they wanted to look at, whether that was alcohol, sugar or fat. but I had done something similar in a prior viz of mine and I kind of called back to that and I thought this would fit really well in the dashboard. Just, you know, to your point histograms take up a lot of vertical space. So kind of a workaround here that I found.
Fiona (39:03)
There's a few things that I really love. for starters, know, there's a big movement at the moment to go alcohol free on things, or at least there is here in Australia. It's in the past it was quite a, people would almost be shocked that you weren't drinking alcohol. And now it's becoming quite the lifestyle choice. And by having the optionality for people to say, I want a virgin or I want a double, like it really makes it,
Kinsey (39:20)
Mm-hmm.
Fiona (39:29)
applicable to everyone and everyone always likes to make things about me. So I think that that really works as well. I really love the balance that you've got in your visualizations. So the way from coming down from even your title and the way that you have the balance in the title, then having your parameters or filters there to drive the visualizations just with those.
Kinsey (39:35)
Mm-hmm.
Fiona (39:56)
two white lines to balance it out on that dark background. It's super eye-catching, the way that you're using a lot of the separator lines as well, the sort of dark navy colors coming in. It makes it visually appealing to want to sit and to interact with the visualization. I think, you know, the other thing, the way that you've taken
the nutrition label and use that as the basis for how you're weaving through that designer aesthetic with those breaker lines coming through. It has a real graphic design theory behind it and the approach.
can you talk me through what your...
influences have been and how you've learnt Tableau and obviously picked up the design of it too.
Kinsey (40:53)
Yeah, I will say this is the first time that I had leveraged Figma in working with Tableau. And I think as I look back on my older vizs, specifically the coffee calculator, I look at that, it pains me a little bit to look at it just because the use of white space was so limited. I think that was back in the day when I was using tiles instead of floating things. So I think kind of the first
big change that I wanted to make with this dashboard was to give it more space to breathe. think that immediately just gives it so much more. It just looks a lot higher end when you give your dashboard space to breathe and you kind of give your user space to read to and kind of see visually how things are supposed to be grouped together. So I know that's not a direct answer to your question.
I think when it comes to inspiration, my sister is a graphic designer. And while I don't have training in the graphic design space at all, I have gotten to see a lot of the work that she has done through the years. We actually worked together at Starbucks. She was in the graphic design studio and I was on the data science team. And I think just having.
personal connection with somebody who was in that space, I paid more attention to the work that she was doing and the work that came out of the graphic design studio at Starbucks. And I think that might have actually influenced a lot of my design choices. Starbucks really goes for that like font hierarchy and like the darker colors as the background. I think I inadvertently absorbed some of those design choices.
⁓ but it's kind of funny because I was sort of dictating my design and deciding what it would look like as I was going through my Figma exercise. one of the things that I was struggling with early on in the viz was the use of color. And I really like when color has meaning. And that was something that I really wanted to come through in this dashboard to not overuse color, but to use it really when it.
meant something in particular. so throughout this dashboard, I use ruby red for alcohol, this yellow for sugar, and then this light teal for fat. And something that I struggled with at first was these were not the buckets that I originally intended for this comparison section. I actually really wanted to have calories. It was kind of like the starting point. But I had built the rest of
the viz and this was kind of the one area where that wasn't aligning with my color scheme. And it was breaking everything. I would look at my viz and I was like, this is so ugly. I can't stand how any of this is looking. And so I did make some data choices to kind of lean into the design choice around the color that I had selected. so I think sometimes there are trade-offs in terms of
what you show and kind of how you use color and how you use design to show those things. And that can impose limitations, but I think overall, the readability of the dashboard benefits so much from the intentional use of color that it was a trade-off I was very willing to make. I don't think I am missing having calories on here to the extent that it would have made sense to add another color. But I essentially kind of started with
poking around all of the different default color palettes in Tableau to just give me a starting point because I know that those have been designed with accessibility in mind and just kind of making sure that you have colors that are distinct from one another. So I kind of started there and then it ended up adjusting, I think it's called the Jewel Brights palette to grab my alcohol, sugar, and fat colors. And then I had gone on to, it's a really
popular website. can't remember the name of it, but essentially like a color palette generator. So I put in my, do you know what it's called?
Fiona (45:07)
coolers.
Coolers.
Kinsey (45:11)
Yeah, pretty,
pretty intuitive. I went onto that website, kind of put in my three colors that I had chosen. I knew that I wanted there to be a dark background, so kind of poked around until I got that. And then I wanted to do the background cards in Sigma too, just to give more, I guess you would call it kind of a visual hierarchy to the dashboard without having to do solid fixes.
around the components just because I was doing that in my working kind of in like as I was working through the dashboard and it just looked really boxy and harsh. And so when I finally jumped into Figma, I was really amazed at how easy it is to use. I don't know if you two use Figma or if maybe there are any listeners who have felt intimidated by it. is the simplest user interface that makes it so easy to do all of your spacing. So I was
Really thrilled to discover that in my last few days of Iron Fist timeline, because I'm not sure what I would have done if it had been really challenging. But that was essentially kind of how I landed on the design. And as I kept going, I didn't want to add too many design elements because, this competition really is about Tableau and using Tableau. And I really wanted the tool to be able to shine, but it just felt like there were some really good ways to.
add some visual separation using Figma. So for instance, in the small multiples chart, I wanted to visually separate the different small multiples without adding lines or something that would just be kind of like visually distracting. So I went through and I mimicked the corners and the structure of these cards that I had added behind my text and my charts and just kind of mimicked that for the small multiples.
I thought that really brought together the visual. looked really messy before I added these in. And then as I was looking through the viz, found I was bothered by how this recipe card was looking at just like kind of looked like a lot of different separate elements and I wasn't sure how to bring it together. But then I borrowed from my design down here with the light background cards and dropped that in here. I didn't want it to be super high contrast.
but just wanted it to be something that would help it feel more visually cohesive. those are kind of the things I was thinking of as I was designing.
Sarah (47:37)
good. I do really like the fact you've used some really bold colors, but because you've given them space to breathe, they don't come and hit you in the face. And I think having that dark contrast behind helps you almost rebalance. And something else that I really like is how you've done the legends here. So quite often, when we do legends, and we're using like that teal color, it's really hard to just
use the color itself for the text because it wouldn't pass any kind of accessibility tests. So I like how you've carried that through there and further down as well.
Kinsey (48:16)
Thank you. Yeah, that was a very last minute addition and I was kind of like one of those 11th hour edits I was making. I like, I really hope this works because I don't want to put a whole legend on the dashboard. You know, it's kind of like, you don't want to take away from the artistry when you're just having to drop in a clunky default legend. So that was, it was a happy, happy surprise when the contrast worked out on the page.
Fiona (48:43)
Yeah, awesome. Well, Kisley, it's been absolutely amazing to have you part of our Iron Viz special. Thank you so much for coming on and sharing some of your secrets and also the journey that you went on to create this most beautiful visualization. I highly recommend that people...
Check out your Tableau Public profile, give you a follow as well, and maybe even connect with you on LinkedIn. We'll make sure all of that information is in the show notes. Thank you so much for being with us today. And now we're gonna pass along to the next person. Thank you very much for being with us.
Kinsey (49:23)
Thank you so much for having me.
Fiona (49:28)
And last but not least, our final guest today is Louis Yu. He's a first year Tableau visionary, but he's been around the Tableau community for a long time. He's a three time Tableau ambassador and the creator of 8 Viz of the Days. He also co-hosts the most loved community events in Singapore, including the Singapore Tug and he does his
virtual community efforts with Games Night Viz. But that's not all, he's also a podcaster with Secrets of the Viz. And so it's a really amazing thing that we've managed to get him on board to talk about some of his things, his secrets on his Iron Viz submission. He's one of those rare people who can entertain, educate and inspire all at once. Welcome, Louis.
Louis (50:20)
Thank you for having me. I'm really excited to be here.
Fiona (50:22)
Yeah,
I'm sure it's nice not to be the host for once and to be the guest.
Louis (50:28)
Yes, definitely.
Sarah (50:31)
So Louis, as we jump in, I'd love to understand the thinking behind your Iron Viz submission. What deliberate design and storytelling choices did you make to bring the concept to life?
Louis (50:45)
So funny story, I think when the theme came out, food and drinks, like many in the community, it wasn't an exciting topic that people were raring to go. But I tried to see what I could work with. There were a couple of topics that I explored with my wife, but none of it was speaking to my true self. I'm, an NNN, a video game. ⁓
kind of nerd. So I wanted to see I could spin an angle of video games around food and drinks. At the start, when I was looking through the data, it felt like I was forcing the topic. But as I dug deeper into it, I felt that there is still a story to be told there. And that's how I started it. I think in terms of storytelling, for video games, it has been around for 50, 60,
years, the industry has gone through tremendous evolution in terms of technology and also preferences from people because what we used to enjoy as games in the past, Pac-Man, Tetris, just name a few, compared to the games that we have these days are very, very different. So the representation of food
in these games has also evolved over time. So I wanted to tell that story of how it evolved and how the food and drinks are actually represented in video games one of the pivotal points in my analysis when I was going through the data was that
the very first game that had food and drinks is actually Pac-Man. So if you think about it, Pac-Man goes around the maze, eating up the dots. You can eat a banana or a cherry, and it gives you powers to eat the ghost. So that was the very first food representation in video games. And as we go on, there was this huge video game crash in 1983 where, I mean, this is evident in many industries. Whatever works, people try to copy.
From 1980s to 1983, many companies were just copying the blueprint of Pac-Man, putting out a new skin for it and just putting out the games. And people were getting sick of it, which from a billion-dollar industry, it crashed to a million-dollar industry, which was close to a 90 % drop. And it took a couple of years before the video games industry actually recovered.
And when it did, we had technological advancement as well. So it kind of moved that needle on how we can actually add more things to games in terms of storytelling, in terms of gameplay, and also representation in food. So that actually formed the basis of my storytelling because I needed a kind of like a glue to stick from the start to the end. And when I was doing the analysis, I think
One of the difficult parts that I go through was that I had three distinct ⁓ elements or stories in my verse, and I couldn't find a way to link them together. So I chatted with a couple of folks in the committee. ⁓ Chris Laudable, one of it, he told me that I needed a anchor to go from one point to another so that
there is kind of like a narrative flow. So for me, Pac-Man was that narrative flow. ⁓ So in terms of design, I lean heavily into the retro vibes of it. I can share my screen and kind of just talk through how it looks like.
Fiona (54:33)
I'm sure everyone's like really waiting to see this visualization, because you did such a great introduction to it and you're like, where is it going to go?
Louis (54:40)
Thank you so much.
So when I first started it, because it being a food and drink, I wanted a of like a design motif to anchor the whole this. And I looked around like, what could I possibly represent food and drinks? Because food and drinks in video games looks very different from real life food and drinks. So I wanted a playful kind of way to show this. So which is why I had this pan.
and ⁓ induction cooker, if you will. And I actually initially wanted to draw this via map layers, but it was too much work. And also the gradient will kill me. In order to draw a gradient, just for the audience sake, you need to draw multiple lines and stack them on top of one another. And each line, can actually have a different color from start to end. So that's how you can draw a gradient.
But there was too much work. There was a lot of data densification. And I had a lot more things to do. So I opted to do this in Figma instead. So the title of the Viz actually came very naturally to me because video games were actually playing with food in the video games itself. So one of the hardest thing when it comes to data was actually classifying which games fall under which category. So I have here three broad classifications of
how food is represented in video games. So food as a goal is games like, know, Pac-Man, where the end goal is actually eating, right? Pac-Man eats dots and stuff like that. Food as a resource is when food is a secondary ⁓ ability or secondary resource within the game itself. So for example, when you're playing games like Super Mario, right? You eat the mushroom, you gain an ability to grow bigger, you can, you know, jump higher and stuff like that.
And then lastly, you have food as a process where you talk about the actual cooking process being represented in the games itself. So if you think about the more recent ones during COVID days is like overcooked and animal crossing where you can actually fish with your avatar, or you can actually cook the dish and you see the pan and the fire, you drop the ingredients in the campfire and stuff like that. So those are games with food as a process.
Unfortunately, there is no data set on the internet that has this classification. So I had to manually go through about 16 to 17,000 games and classify them manually. had some help from ChatGPT and also from my experience from playing games, because I've been playing games since I was young. So I do have a lot of knowledge on what, food is represented in games. So that helped a lot.
that being said, I managed to classify all 16,000 games. So you can see from here, this is a radial chart. A Coxcom chart that actually represents each of the different pillars. You have foot as a resource.
food as a goal, food as a process, and also games with no food mechanics. So this was particularly hard to build because with La Data Viz tools, you have the CODSCOM chart that you can actually create and then export it in Tableau Desktop. But I didn't want to rely on tools. I wanted to challenge myself as well because Ion Viz is also a chance for me to kind of tackle problems that I don't usually tackle.
And one of the struggles that I had this year was map layers and trigonometry. So I wanted to really dive deep into those and try to upscale myself. So I went in all, did all this breakdowns of all the radials. And also I had this that basically shows you each of the games that were classified within the top rated games. So if you hover over, you can see the game titles and
what categories is being ⁓ grouped under. So this was kind of like the, I would say the opening piece that I wanted to give people with to start off their journey on looking at games, the video games industry. And if you flow down here, this is my main centerpiece. Because like I said, I wanted Pac-Man to be the motif of the entire Viz. So...
I intentionally created this as kind of like a arcade machine, like a retro arcade machine where you see, know, Pac-Man eating up the dots. And this was initially planned to be kind of like a curvy timeline, but I felt that with curvy timeline, it's very hard to annotate. So that was my biggest challenge when I was building this chart, because I can't actually annotate many points and have long text within, because you can see like this
bunch of text over here takes up a lot of space. So what I opted to do was to add on to like the interactivity that you would get from an arcade machine. this actually came from a feedback from my last year's IMD submission. So last year I had this television screen with a game console in front and can actually swap the cartridges to change the visualization.
But the feedback came back to me was that ⁓ they had a lot of fun with the cartridges, but the interactivity of that didn't flow through to the rest of the chart. So it's only at the top. So I thought about that hard, which is why I had this part where you can actually click and switch the dishes on the pan itself. And I wanted to bring that interactivity throughout the whole viz. So within this arcade machine,
you have this little forward and back button that allows you to go through the different story points of the history of video games. So we start off with Pac-Man in 1980s. If you click, you'll go through to 1983, where you see the video games industry crash. And then you see where it actually revived in actually 1986. And then you can actually click through.
all of these different points to see the different annotations because it's a lot of information that I want to share but I wanted to do this in a fun manner where people can actually click through. So this is kind of like the overview of the history but you can also look at the individual history of the different peelers. So obviously how food has evolved from as a goal is different from how food has evolved as a resource in video games. So if you can
You can also click on these different buttons and it changes to the different story points so you can actually go through. So this took me probably the longest time to build. And then the last part, interesting enough, I had a lot of iterations as well. So this was the very first version, right? So I had this piece where it's meant to be exploratory, but I wanted to also give people kind of like a too long don't read version because
Not everyone has that patience to click through so many buttons and all, even though it's meant to be fun for them. So I actually had this pretty long paragraph of text, that I kind of sum up the entire history of video games and how food is represented, followed by, a chart, a network chart of how realistic food or how diverse food has become.
in video games as taking as a case study for Final Fantasy 15. So they had like 115 dishes in there and equally amount of ingredients as well. we hover over, this was what I wanted to do to show how realistic the food actually looks in game.
But you can actually still explore this by clicking on a recipe to see where the ingredients are coming from. You can also click on the ingredients to see what recipe it fits into. But this was really long. And I felt that it is not accessible for many people who are just going through this entire viz. So what I ended up doing after lots of iteration was to break this
into three different eras of the evolution of food. So the first evolution was when we started to have the cooking process in Virtual World. So you have games like Cooking Mama during the release of like Nintendo DS back in the days. So Nintendo DS comes with a stylus, right? that they had this cooking game
that you could actually use the stylus to imitate cooking processes. So you can slice onions using a downward or a straight line kind of action. You can tap to chop finely. You can do a circular motion to actually stir the pot and stuff like that. So it became very immersive as a video game. And then we move on to the second era, where cooking became a subplot.
of the main game, right? So you have games which are very focused in narrative. So we call them the role-playing games. And role-playing games usually, you take the role of an avatar, you go explore the world, and food is never kind of like the main topic. But ⁓ Suikoden 2 actually had a subplot where you could actually participate in a cooking contest, right? So the cooking contest has like judges.
and each of the judges has different preferences to different foods. And they will rate the food differently depending on what you present to them. So it's really interesting. So what I did was that you have this little button here that when you click, goes to the game footage over here where you can actually see how it looks like in the game,
Yep, there we go. So this is actually how it looks like. So you have a cooking contest where you have the judges sitting at the back and you prepare the food accordingly. So it's a very fun way to incorporate food into the games itself. ⁓ And that was the second era. And then the third era is when food became really realistic in the game itself. So Final Fantasy XV was one of the first games that did this.
the studio actually hired a team of chefs to go out into the open and cook over hundred dishes and they took a lot of photos of it for the game designers to recreate in-game so if you hover over you'll see all the different ⁓ food items and you can see that it looks really realistic because that's how they actually based it off I actually have this I found this article which was really interesting and it showed
the behind the scenes of the chefs actually cooking the dish. And this is how it looks like in real life, and this is how it looks like in game as well. So yeah, it's really amazing how technology has allowed us to be able to represent diverse food and also how real it looks like in game. So I ended off my video by saying that the
industry works in circles, right? We've seen that happen a couple of times and with the rapid developments of virtual reality and augmented reality, we might be able to play with food without actually playing with my playing with food because you can actually, you know, put on the goggles and, you know, cook in virtual world and stuff like that. So that's basically the entire story of my viz and you know, I've gotten some feedback that people really enjoy.
going through this verse because some of the games that they didn't think had food actually had food in them and they went back and they reminiscent on some of the elements within the game itself it really brought them down memory lane so I'm happy that ⁓ this resonated with a couple of folks
Fiona (1:07:15)
Gee, there's so many elements in here to dive deep into, but starting at the beginning, which was in your research, how long do you think that it took you to get all of the data that you needed?
Louis (1:07:30)
So I had some help because last year we for games like this we created a starter kit for anyone looking to kind of practicing visualization using video games data. So I had that catalog of video games which was the 16,000. The only thing I needed to do was to update with 2025 games which took me like maybe a couple of days because I needed to
fetch out ⁓ the genres, the platforms, the date of release, the description, and the titles and stuff like that. But it wasn't too hard to get through that. The other part was, again, classifying them into the three pillars, which was there's no data out there. So.
Fiona (1:08:17)
Yeah, I think
it's a common story that a common thread that we've found as we are interviewing people, the links that they've gone to to get the data and to research it as well. What recommendations do you have? You know, we were talking about the volume of people who had put submissions in this year was perhaps a little less than other years that we've seen. ⁓
One of the things that might hold people back is actually going out and doing that research in order to tell us a good story and to get the data. What are the things that you recommend in terms of harvesting data to do activities that are fun, whether it's games, like this or iron vis submissions?
Louis (1:09:08)
think the one thing you could do, I we live in a AI driven era now, it or not. And I think people shouldn't shy away from using AI tools to help you organize your thoughts. I mean, you might have an idea, but you don't know how to go around researching on that topic itself. So one of the initial tasks that I gave ChatGPT was, help me do a deep research. I have this information, Pacman.
was the very first game that had food in it and in modern games I have like role-playing games like XYZ that had food in games where you can drink to restore your health and stuff like that so I gave ChatGPT a lot of context behind how food is represented in video games and I told them that can you go scour the internet and see if there's any link to show how it's evolved over the years
like with any AI output, always double-check on what they give you because they tend to hand-listen it.
They tend to invent things that are not there, but thankfully because I come from the video games industry I know when they are talking rubbish. It's like no that didn't happen at all So I went through it. I found a couple of threads that I could continue so I Fed it back to them because because ChatGPT at the end of the day It's also kind of like an information bank. Whatever you feed them You get
extrapolation of that. if you feed them good data, they'll give you good data and stuff like that. it went through, I think it took me like an entire week of back and forth with the AI just to get a thread of a story that I was going for. And then once I had that, I got to preparing the data to tell that story itself. So that took maybe another two weeks.
to prepare the data and write out the narrative that I wanted. And then the last two weeks was spent designing and creating all the charts.
Sarah (1:11:14)
Wow, what a process. It's so amazing. And it's been great to hear and learn so much more around video games and its connection to food. Louis, thanks so much for coming on. ⁓ We look forward to listening to your podcast, which you're doing also around Iron Viz So yeah, thank you for giving us the time today.
Fiona (1:11:14)
Yeah.
Louis (1:11:35)
Thank you for having me. It was so great. Thank you.
Fiona (1:11:37)
Thanks, Louis. It was awesome.
Sarah (1:11:43)
And that wraps up our Iron Viz special. What an absolute showcase of creativity, courage, and community.
Fiona (1:11:51)
I mean whether they make it to the finals or not, every person who enters Iron Viz helps push the craft forward, especially when they do something like this, take their time out of their day to come and share some of their insights. That's something to celebrate.
Sarah (1:12:07)
So a massive thank you for everyone who joined us and the Data Fam from cheering them on the sidelines.
Fiona (1:12:13)
And if you've loved this series, hit like, follow our podcast, subscribe, and please share it with anyone who wants to know the secrets behind Iron Viz or just Viz Success.
Sarah (1:12:25)
This is Undubbed, where we're unscripted, uncensored, and undeniably data from your friends at Dub Dub Data.
