Become a Creator today!Start creating today - Share your story with the world!
Start for free
00:00:00
00:00:01
Episode #46: Makeover Monday
141 Plays9 years ago
Welcome back to the PolicyViz Podcast! I’m pleased to welcome Andy Cotgreave and Andy Kriebel to the show this week. Each week, The Andys, as it were, are the primary authors of the Makeover Monday project in which they–and dozens...
The post Episode #46: Makeover Monday appeared first on PolicyViz.
Recommended
Transcript
Introduction to Georgetown's Summer Executive Institute
00:00:00
Speaker
This week's episode of the PolicyViz podcast is brought to you by the Summer Executive Institute at the Georgetown University McCourt School of Public Policy. The McCourt Executive Institute offers short courses that are specifically designed to enhance key skills.
00:00:15
Speaker
Small classes and hands-on projects allow you to engage with expert faculty at Georgetown in a format that is convenient for busy professionals. To learn more and to register, please visit mccourt.georgetown.edu slash exec-ed slash shortcourses. Enhance, energize, and expand your professional skills this summer at the McCourt Executive Institute.
Meet the Guests: Andy Cotgreave and Andy Kriebel
00:00:49
Speaker
Welcome back to the Policy Viz Podcast. I'm your host, John Schwabich. Two very special guests this week. I'm joined by the Andes, Andy Cottgrave and Andy Kreebel of the Makeover Monday Project, along with many other things. Andy Cottgrave, technical evangelist at Tableau. Welcome. Hello there, John. How are you? Very well, thank you.
00:01:07
Speaker
Back on the show, second time. Yes, a returnee. Andy Kreebel, Head Coach at the Data School in London. Andy, welcome. Thank you, John. Thanks for having us. I've got the Andes here. Let's just jump right into it. You guys have had this project Monday Makeovers for how long have you been doing it now for? 18 weeks. I've been doing it for a long time, but Andy jumped on the bandwagon 18 weeks ago. He's just following. Yeah, so Kreebel, you've started what? I mean, you've been doing this for eight years, right?
What is Makeover Monday?
00:01:35
Speaker
Yeah.
00:01:37
Speaker
November last year, I was just trying to use Tableau for some just internal company analytics. And I was like, Oh my gosh, in my job, I'm not using Tableau anymore. And I realized I'd forgotten some basic stuff. So I was like, Oh, I need to get back on the wagon. And so I pledged to Andy. I said, look, why don't you and I just each week, you know, I'll do a makeover with you. We'll find a chart. We'll find a chart that needs a makeover or could have a different perspective taken on it. And we'll both blog our experience each week.
00:02:06
Speaker
And, you know, my motivation there was, you know, it'll be a great way for me to use Tableau every week. And then it kind of ran away with itself and it's become, it's become a pretty huge sort of social data experiment in a way, because what's happening is Andy and I, well, Andy prepares the data each week, gets it published. And we say to the world, right, Andy and Andy are going to make over this job, but what else can you do with this? And it's honestly been amazing. We've had 613.
00:02:35
Speaker
different makeovers in 18 weeks, which is about 34 people a week, taking the data, taking a chart, making it into something else, finding new stories, flexing that data with skills and learning from
Community Engagement and Participation
00:02:48
Speaker
each other. It's just, it's honestly, from what I expected it would be, it's become something amazing.
00:02:54
Speaker
OK, yes, so like Andy said, it's been it's been an amazing community project. I was really excited that Andy decided to join me on this, although it did put a bigger constraint on me requiring me to do it every week. So I just give him the data and he goes off and does all the easy work.
00:03:10
Speaker
But it's what's been really amazing to me. Like Andy said, we didn't know how the community would respond to this. And, you know, it's just been it's been amazing. Our numbers are slightly different. I have the official numbers and I'm at six hundred and one. But, you know, tomato, tomato, potato, potato. So but, yeah, it's been we typically see there's probably about I would say what what 15 or so people that do it every week. Maybe something like that.
00:03:40
Speaker
Yeah, and then a bunch of people dropping in and out. We still are getting new people trying their first time for Makeover Monday. It's just amazing.
00:03:51
Speaker
I think one of the benefits of providing people the data in a nice consumable format, I provide it both in Excel and Tableau data extract format so people can just connect to it right away and get going. They don't have to worry about any of the data preparation because Andy and I want people to just use Tableau or use whatever tool it is. I mean, we've had this one person keeps using Illustrator, it makes us all look bad.
00:04:17
Speaker
because everything's just so beautiful. It's not that much work to prepare the data, and I think it's really worthwhile doing it that way.
Choosing Charts and Organizational Tools
00:04:26
Speaker
What are you looking for when you're looking for a graph or visualization to makeover? Oh my gosh. Well, that's really the easy part.
00:04:35
Speaker
If I look at a chart and I say, what the hell? Then I just tag it in pocket. So Andy Cockrave actually introduced me to pocket a while ago. So that's been an essential resource for this project. Because I use Feedly for consuming blog feeds.
00:04:54
Speaker
and I can easily just see something in there and tag it as a makeover in pocket. Then I have this repository of things that keep coming back to. Andy and I share an Evernote document that has a list of all of the ones we're going to be doing and all the ones we've done. Technically, he and I could cheat if I have the data ready. He really didn't have an excuse this Monday for being late because the data was ready early.
00:05:22
Speaker
I was away this weekend. Oh, there it is. He does have an excuse. It may not be a good excuse. Right, right. But yeah, I mean, it's been I think just we're trying to make it so people can just get going right away and build something. And I think what Andy and I both like is when somebody takes the original visualization and really tells a much more compelling story with the data. And that's really what this is about is
00:05:51
Speaker
Sometimes the visualizations are good and sometimes they're terrible. But it's more about how can you tell the story differently. That's the source visualizations.
00:06:02
Speaker
I was going to say, we're trying to find charts to make us question what the people were thinking when they were designing them. But again, sometimes we do pick good charts. If the data is an interesting story, then how else can you retell the story? Or can you even find a different story than the one the original designers were trying to come up with? Because we're not trying to
00:06:24
Speaker
sort of go, oh, you made an awful chart, and aren't we all better, right? You know, that is explicitly not what this is about. It's just like, okay, you made a chart, how can other people remake it and come up with something new?
Critiquing Visualizations: Principles and Challenges
00:06:37
Speaker
And, you know, again, what we're trying to do in all our posts is say, but if we're going to talk about the original chart, say, well, this is what we liked, and this is what we might have improved. And then here's a different perspective on it. And when you have up to 30 people a week doing
00:06:51
Speaker
these different versions of the chart, you see incredible new perspectives. It's amazing because we're focused on a massive diverse set of data sets. Gender, urban diversity, slave trade, there's a bunch of football stuff and soccer stuff and sport data, police violence. Some really fascinating data sets that this community, as it grows, is finding
00:07:15
Speaker
amazing new things in this data. Yeah, that's that's really interesting that it's not just on critiquing a graph. It's sometimes just taking a different perspective. But I'm curious on the ones in which you are critiquing saying this is not a great graph. This is how we might do it. How do you view your responsibility as the person doing the makeover? How do you feel like you need to responsibly critique and and sort of expanding on that? How do you feel like the feel that self of data visualization is doing critiquing other people's work?
00:07:42
Speaker
So I can go first, Andy, and talk a little bit about why I started this in the first place. So I started doing these makeovers a long time ago because I wanted to use it as a learning exercise. And I was heavily influenced by Steven Fuze books. You know, in a lot of his books, he takes a visualization and explains, you know, why doesn't it work well?
00:08:02
Speaker
Well, I wanted to sort of go a step farther and say, okay, well, here's what works well. And here's what doesn't work well. Because sometimes the chart isn't totally terrible. And you can tell that maybe the person had a really good title, they chose good colors or whatever it might be.
00:08:16
Speaker
So it's not about criticizing the person themselves that created the visualization because I don't think anybody ever creates a bad chart on purpose unless you're with Fox News. And then it's more about how can we use it as a learning exercise. And what I really like when I see makeovers from people that have participated is when
00:08:37
Speaker
They write a blog post that goes with it and they explain their perspective on what worked well and what didn't work well, because that's really where the learning comes in. Just taking the data and building something new, I don't actually think it adds value, but it doesn't add as much value to your own personal learning as it does actually trying to understand the decisions that the person made and what you yourself might do differently. I think that's really where the learning comes in. Yeah.
00:09:02
Speaker
And I agree with everything Andy said there. And then one example that we've had was when in week nine, we remade a donut chart from the Daily Mail. It was all about wages of football players in the UK, soccer players in the Premier League. Right now, donut charts, clearly alarm bells getting rung, the Daily Mail aside that does not fit with my political leanings. So my hackles are up right as we first start looking at this chart. And
00:09:31
Speaker
It was, well, flawed, right? So I, in my critique that week, I wrote, my first opening paragraph was just aggressive and probably because of the type of chart and the, well, in fact, definitely because of the type of chart and the site it was on. And of course, what happened was that Nick from Sporting Intelligence, who was a guy who wrote the article and created the chart, then got in a bit of a huff on Twitter. And we ended up, you know, with quite a lot of back and forth, you know, it was a bit of a, it was an unnecessary argument, really, because
00:10:00
Speaker
had I looked to our principles and said, the data's great, this bit of the donor chart works well, and here's what I would have done to improve it, the whole argument would have gone a different
00:10:12
Speaker
route. I didn't stick to the principles of my bike that week and felt the impact of that. But then when you widen it out, we've talked about this previously on this podcast, when you widen it out, it's if you criticize anybody in any field, and it happens in date of years quite a lot, if you criticize them aggressively, then you will be met with aggression and back and
00:10:35
Speaker
It just is not a constructive way to advance this field, I don't think. I mean, do you think that's always the case, Andy Crebel? I guess I have to use last names too. You mentioned Stephen Few, which of course, his many posts that have that sort of have a tone to them. It's a bit biting, it's a bit aggressive.
00:10:54
Speaker
Do you feel that he's sort of the outlier in the field and that most are a bit more constructive? Or do you sort of wish that people would, you know, maybe tone it down and really focus on the presentation of the data, not so much on the personalities or on the tone? Yeah, I mean, I don't think I've ever seen Steven attack a person. You know, he attacks the visualizations, which is similar to what Andy and I are trying to do. And attack is probably not the right word either.
00:11:20
Speaker
We're trying to critique the visualization. Steven's books were the first ones I read when I got into data visualization, so I think that's why I'm so influenced by them. I just found it a fascinating way to learn. Really, what I learned the most about through reading his books was the thought process.
00:11:36
Speaker
And how do you take something that's good and gradually turning into something that's better? And we see a lot of that in Cole Nussbomber's book. Her book is an amazing example of doing exactly those same things and telling a much more compelling story with the exact same data set. And that's really what I mean, I'm learning so much through doing these every week. And it's really so much of the value just comes with working, working with different data sets every week.
Educational Applications of Makeover Monday
00:12:01
Speaker
and because it forces you to think differently every week. If you're working with sales, superstore sales every week, you're going to become dull. We're hearing that from a lot of the people that are just doing this for the first time. It's an amazing way for somebody that's new in the data visualization community to get really clean datasets to just practice with. There are a couple of great examples. David Perez, who's a guy based in London by day,
00:12:28
Speaker
He's working with sterile bank data. I think it's in the bank. He's got a very strict style guide to work with. By night, he becomes this blogging, crazy, creative superhero. Through Makeover Monday, he gets to experiment and try new things out. One thing I noticed now over 18 weeks is that we're seeing things evolve. In week four or five,
00:12:54
Speaker
we started doing some slope charts and then a few weeks later lots of people did doing slope charts and then I think Andy did a really tall long a tall and thin dashboard I believe it was on police violence and then for a few weeks people are trying out tall and thin things and then it became small multiples and currently everyone's doing hacks maps on May Cover Monday and so people are trying things out it's amazing to see it
00:13:18
Speaker
And as people, the people who are blogging about their experience as well, you know, you're getting an idea of that process. And it's really interesting seeing people's process of how do they get to the end result as well. And again, that's just one of these unintended consequences where it's like, wow, we've now got this gang of people growing each week who are developing skills and flexing, you know, trying out things that maybe couldn't do at work. Yeah. Crebel, I'm curious whether you're trying this sort of approach when you are teaching at the data school.
00:13:48
Speaker
Oh, yeah, definitely. It's one of the I think the best ways to teach people the first two days at the data school. They actually don't even use computers. I make them use those fat like kindergarten crayons and like giant eight. Is it a four paper or a four? What's the really a two paper, Andy? Is that what it is? Those giants? Yep.
00:14:07
Speaker
Yeah, so and we do just do makeovers by hand. And we talk when we talk through the process and I'm trying to teach them about the data visualization process and tool agnostic things that are just just super, super important to learn before you dive into any tool whatsoever. So yeah, so we do I encourage them to follow these same ideas. And when they're critiquing each other's work,
00:14:32
Speaker
I want them to critique not on the person that created the visualization, but, you know, how can whoever created the visualization in the data school, how can they make it better? Obviously, they think they did a great job when they present, but maybe there's a way to do it better. And that's, it's a way for them to learn how to give constructive feedback to each other. And it's, it's a really good thing that, you know, as transferable, anybody could do this, right? You know, I've done it internally at Tableau now a couple of times.
00:14:57
Speaker
as the Makeover Monday homepage grows, it's going to end up with 52 source charts, 52 source data sets of 52 times N, examples of different ways to makeover the chart. One of the other things I think Andy would agree with this is that the most productive weeks are the weeks in which we have data sets that only have as few as 10 records, 10 or 20 records. Because what happens there is
00:15:23
Speaker
people are focusing on, well, what's the story in these small numbers with such a little thoughts to play with, then it actually frees you to think about how to communicate something creatively.
Storytelling with Small Datasets
00:15:33
Speaker
And when we've done ones with big data sets, you know, a few thousand rows, we actually are getting fewer people getting involved in those weeks. And so it's something, you know, anybody doing any kind of database training or workshop could just pull one of these and see what their group makes of the data.
00:15:49
Speaker
Yeah, no, it becomes a great resource for the community. I'm also curious about the conversation amongst the people who are doing the makeovers. Are people talking about their makeovers and are people critiquing each other's makeovers? I mean, I can imagine here's a donut chart that we're going to make over and someone makes it as pie charts and someone critiques those. Are you seeing those sorts of discussions?
00:16:10
Speaker
You've had one this week, haven't you, Andy? You're in the middle of an argument on Twitter about color schemes. Well, I wouldn't say argument. Yeah, I mean, John, we haven't actually seen a lot of people critiquing each other. I think I know I've been very cognizant of not doing that because I don't want other people to be discouraged from participating.
00:16:34
Speaker
And Andy and I, we talked about that at the very beginning. It's like, we don't care what somebody creates, we're going to include it in our gallery. So I don't know if you've seen the Pinterest board, John, but it's amazing. You can see this incredible variety of visualizations, and I just want people to participate. Really, I don't care what they create. They don't critique mine either. So we could do that, yeah, but I think that would actually be detrimental to this particular project. Yeah, and what's happening on Twitter this week is
00:17:04
Speaker
You know, there's a bit of a conversation about a color scheme on one of the charts. And that's, that's been pretty, that's a great conversation. You know, color is obviously a fundamental aspect of what we do. There was one week where we did mapping police violence. So it was data about when police kill people in the US. And that actually generated quite a lot of conversation because the data had ethnicity of who was getting killed in the US. And then you end up in really interesting grounds where actually you can tell
00:17:33
Speaker
You can skew a story just by the way you design charts, so you're not actually being deceptive, but you're leading people down a particular opinion. And so that got a fairly good amount of debate. There was another one, actually, Steve Wexler got involved. Which one was that in? It was saving. There was one we did, something about savings, right? There was a chart about where people in America and the US had been asked, how much do you have in your savings account?
00:18:01
Speaker
from which a story in a chart was spun about, oh, Americans are not saving enough. And Steve Riley pointed out that most of us then went and just kind of remade the chart without actually looking at the question and thinking, well, you know what? Just because you've got nothing in your savings account doesn't mean you don't have much savings. So again, that conversation that week was around, well, hang on.
00:18:25
Speaker
don't just remake the chart, think about the story and the data and is the actual fundamental question the right one, which was a good wake up call because that was week four. So I think we've adapted that criticism as well. Yeah. And this week, so we're starting to see a lot of more people designing for mobile, John, um, I think primarily because I think that's the way most of us consume these visualizations and Twitter now. So we're seeing a lot of people trying to design things. Obviously most of the people that are doing it are doing it in Tableau.
00:18:52
Speaker
And we're finding lots and lots of problems with designing visualizations for mobile devices in Tableau, like filters are really, really difficult to work with and parameters are really difficult to work with. So it's providing us an avenue to get feedback to the developers so they can make the product better.
Designing for Mobile in Tableau
00:19:08
Speaker
So it's in turn providing Tableau a testing platform and a feedback platform for trying to make the product better and better and better, which is fantastic.
00:19:16
Speaker
Yeah, that is really interesting. So it's interesting you talked about, you know, some of the lifecycle of the actual data as opposed to just doing the charts. I'm also curious about whether people are talking about the technical aspects of building things in Tableau or people are submitting talking about how they built certain things or asking others how they built different dashboards or visualizations.
00:19:34
Speaker
There's a little bit about that. I've certainly had a couple of blog tips, you know, that some of the things I've done, I'll be like, Oh, that's, I've actually done something kind of cool here. Just sort of access labeling or font positioning stuff. And you're like, well, you know, they, they can become good blog posts. Andy Krieble is very good at that.
00:19:50
Speaker
Yeah, I don't know. Maybe we're not seeing as much of that as we could do. What's interesting, it's certainly not something where we dictate people should use Tableau. As Andy said, someone's using Photoshop. We've had one person do a bunch of hand-drawn makeovers. In fact, we'd like to see more people using different tools getting involved. We've tried to encourage people from other communities of different vendors and different tools and different languages to get involved. In fact, we can make that call on this podcast. Whatever tool you're using, listener,
00:20:20
Speaker
come and get involved in Makeover Monday. And tell us about your process. Yeah.
Get Involved with Makeover Monday
00:20:24
Speaker
Well, it's great. It's a great project, and I've enjoyed following it. And good luck the rest of the way. Thank you very much. Thanks very much, John.
00:20:31
Speaker
Andes, thanks for coming on the show. This has been fun. And thanks, listeners, for tuning in once again. Please let us know what you think about the Makeover Monday project. And as Andes said, please join their conversation, download the data, and make your own visualizations in any tool, not just Tableau, any tool. So thanks again for listening. Until next time, this has been the policy of this podcast.
00:21:02
Speaker
This week's episode of the Policy Vis podcast is brought to you by the Summer Executive Institute at the Georgetown University McCourt School of Public Policy. The McCourt Executive Institute offers short courses that are specifically designed to enhance key skills. Small classes and hands-on projects allow you to engage with expert faculty at Georgetown in a format that is convenient for busy professionals.
00:21:25
Speaker
To learn more and to register, please visit mccourt.georgetown.edu slash exec-ed slash shortcourses. Enhance, energize, and expand your professional skills this summer at the McCourt Executive Institute.