Jon Krohn: 00:00:00
This is episode number 651 with Kate Strachnyi, data community entrepreneur, and author of the new book, ColorWise.
00:00:11
Welcome to the SuperDataScience Podcast, the most listened-to podcast in the data science industry. Each week we bring you inspiring people and ideas to help you build a successful career in data science. I’m your host, Jon Krohn. Thanks for joining me today, and now let’s make the complex simple.
00:00:42
Welcome back to the SuperDataScience Podcast. Today, I’m delighted to be back on air with the wonderful Kate Strachnyi. Kate is a multi-time data science book author. Her latest book was recently published by O’Reilly and is called ColorWise. It’s a beautiful comprehensive guide to the effective use of color when communicating data visually. She’s also the founder of the DATAcated Circle, a community of data professionals committed to engaging and learning together. She is a megastar on LinkedIn, where she has over 170,000 followers and was twice recognized as a LinkedIn top voice for data science and analytics. She’s also big into long-distance running. Her longest to date was a 50 mile. That’s 5-0 mile, Ultra-Marathon in New York’s Central Park.
00:01:27
Today’s episode should appeal to technical and non-technical folks alike because I suspect that pretty well any listener of the show presents data and could benefit from learning how to do so more effectively with the intentional use of color. In today’s episode, Kate details why now was the perfect time to write her book on color for data visualizations, why the intentional use of color matters, what thought process you should follow to select a color scheme for a visualization, special considerations for color choice, such as accessibility, cultural understanding, and due to human psychology, she also talks about how to effectively use multiple visualizations together in documentation, a presentation or a dashboard. And she fills us in on her favorite data biz tools. All right, you ready for this vibrant, full color episode? Let’s go.
00:02:21
Kate, welcome back to the SuperDataScience Podcast. Oh man, what a great guest you always are.
Kate Strachnyi: 00:02:29
Thank you for having me back.
Jon Krohn: 00:02:32
So you were most recently on the show in episode number 441. You were one of my first ever guests when I became host of the SuperDataScience Podcast. But you also had one with Kirill at some point before that?
Kate Strachnyi: 00:02:45
Yes, yes. I think a year or two before that, I was on the SuperDataScience Podcast.
Jon Krohn: 00:02:51
Cool. Let’s be honest, let’s tell the audience, who gave you the better hosting experience, was it Kirill or me?
Kate Strachnyi: 00:02:56
Well, I’d have to pick you for one reason.
Jon Krohn: 00:03:02
Really?
Kate Strachnyi: 00:03:03
Listen, Kirill, we had some recording issues, so it wasn’t really him. And then we had to take a break because he had to go take care of something, and then we came back and then we had to re-record another part. And for me, re-recording is a nightmare because I wanted to come out authentic, and if I’ve said it once, saying it again feels kind of awkward and I’m not as excited.
Jon Krohn: 00:03:30
Oh, man.
Kate Strachnyi: 00:03:30
But besides that, it was great.
Jon Krohn: 00:03:30
Yeah, I mean, I thought you were going to give a really funny reason and then you gave a real one, so that was a big surprise for me. I thought it was going to be because I’m bald or something or that I was better. You like all of your podcast hosts to be bald, was going to be a reason.
Kate Strachnyi: 00:03:45
I was going to say the guitar, but I’ve never seen you actually play it. So I don’t know if that’s just for…
Jon Krohn: 00:03:52
Ah, no. I even did once on air. There’s an episode. So the final episode of 2021, episode number 536 ends with me playing guitar and singing.
Kate Strachnyi: 00:04:09
Okay, well I’ll have to check that one out. That can give you bonus points there.
Jon Krohn: 00:04:10
Yeah. And actually, I wanted to end the 2022 episode in the same fashion, but I broke my finger and it’s like…
Kate Strachnyi: 00:04:18
I mean, is that a real excuse though?
Jon Krohn: 00:04:21
No, I only know one song, so I deliberately broke my finger.
Kate Strachnyi: 00:04:25
What’s the song you played in 2021?
Jon Krohn: 00:04:28
Ooh. I played a song that was made famous by Nirvana in their Unplugged in New York performance, and it’s a song that’s by the Meat Puppets. I’m getting really close. It’s called, Oh Me. It’s called Oh Me.
Kate Strachnyi: 00:04:43
Okay.
Jon Krohn: 00:04:44
By the Meat Puppets.
Kate Strachnyi: 00:04:45
Can you sing it? I don’t think I remember the whole…
Jon Krohn: 00:04:50
Oh. Context, Kate. You can go back and listen to the podcast episode. Good try though, good try. I can’t sing because my finger’s broken.
Kate Strachnyi: 00:05:02
Oh, okay. There you go.
Jon Krohn: 00:05:02
Yeah, unfortunately. So, all right, where in the world are you calling in from, Kate?
Kate Strachnyi: 00:05:08
New Jersey.
Jon Krohn: 00:05:10
Awe yeah. I know that there’s a really nice restaurant called Mr. Sushi near where you live.
Kate Strachnyi: 00:05:15
Oh, yeah? Really?
Jon Krohn: 00:05:18
It’s the only place I’ve ever seen where you can get cheese in a sushi roll.
Kate Strachnyi: 00:05:23
Oh, interesting. And another interesting…
Jon Krohn: 00:05:26
It is interesting. I don’t recommend it.
Kate Strachnyi: 00:05:28
Is it like a Philadelphia roll, or…?
Jon Krohn: 00:05:29
Nope, nope. Nope.
Kate Strachnyi: 00:05:30
Because that’s cream cheese.
Jon Krohn: 00:05:31
Yeah, no. Yeah, the Philadelphia roll thing, the cream cheese thing, I think that is a uniquely American thing, but that’s somewhat widespread.
Kate Strachnyi: 00:05:36
Yeah.
Jon Krohn: 00:05:39
Yeah. New Jersey was the first place that I ever had cheese. Hard cheese.
Kate Strachnyi: 00:05:46
Like an American cheese? Like a cheddar cheese? Like a hard Parmesan?
Jon Krohn: 00:05:46
Yeah, yeah. Yup.
Kate Strachnyi: 00:05:47
That’s kind of weird.
Jon Krohn: 00:05:47
Yeah, exactly. It was really weird. And so that’s why I’m saying it’s possible in New Jersey to do but I’m not recommending it.
Kate Strachnyi: 00:05:59
I don’t think I’d order that one, yeah.
Jon Krohn: 00:06:02
Fish and cheese together at last. Everyone’s dream combination. So while we’re not really here to just talk in this episode about fish and cheese together, although we could probably make an entire episode of it, we want to talk about your new book. So we have…
Kate Strachnyi: 00:06:20
I have a new book.
Jon Krohn: 00:06:21
You have a new book.
Kate Strachnyi: 00:06:21
Right.
Jon Krohn: 00:06:21
I have it right here with me. You very kindly sent me a copy. I asked Kate for an electronic version so I could research for this episode, and she sent me a physical copy and it wasn’t like you got O’Reilly to send me one.
Kate Strachnyi: 00:06:37
No, no. So O’Reilly sent me 20 copies when the book came out, and I still have some back there behind you, if you turn around, you’ll see them. You’ll see them all.
Jon Krohn: 00:06:47
Behind you too. Why did you store them here in my apartment? Why did this even happen? Who let you in?
Kate Strachnyi: 00:06:56
So yeah, I had some copies and I figured it’d be quicker to get it to you in time for the podcast.
Jon Krohn: 00:07:02
Oh yeah. I really appreciate that. Yeah. So yeah, I noticed. Yeah, it seemed like it had been sent just from your Amazon account. Thank you very much. And it’s a fantastic book, I actually already wrote an Amazon review about it, gave it five stars, almost gave it four… No, no, no. It was obviously a five-star review. It’s exceptional. The amount of research that you put into all of the topics, just tremendous. I wasn’t sure when I was thinking, “How much is there to write about color?” There’s a ton. And for me personally, one of the most exciting things was that it starts with the neuroscience of color perception, and my background is in neuroscience, and so I loved getting a refresher on that. I thought it was extremely well done, and unsurprisingly, given that it is a book about color, it is filled with fantastic figures, full color figures, throughout. I think just about every page has a full color figure in it. Yeah. So it also makes her a very easy read. It’s almost like a picture book.
Kate Strachnyi: 00:08:06
It is a picture book, yeah. It’s a color book, so…
Jon Krohn: 00:08:09
It’s a coloring book.
Kate Strachnyi: 00:08:10
You can color it in.
Jon Krohn: 00:08:11
The appendix is just in the coloring. You get to color in some charts.
Kate Strachnyi: 00:08:14
I should have done that. That’s a good idea.
Jon Krohn: 00:08:15
That’d be really fun.
Kate Strachnyi: 00:08:18
Oh my God. Okay. Edition Two.
Jon Krohn: 00:08:19
Yeah, you get to practice everything you learned about getting your cards…
Kate Strachnyi: 00:08:24
Yeah, color in this chart. Oh my God. That’s…
Jon Krohn: 00:08:26
You get a free set of crayons with your purchase.
Kate Strachnyi: 00:08:28
No, seriously, you’re giving me a good idea right now. I’m going to have printable coloring sheets for data visualization where people can practice their skills.
Jon Krohn: 00:08:37
It started as a joke, but Kate right now is literally taking notes on air. Just like that.
Kate Strachnyi: 00:08:41
I’m literally taking notes. I don’t want to forget.
Jon Krohn: 00:08:45
So that’s cool. And then actually, if you’re listening to this and you’re listening to it right when it comes out, then you actually have a chance to get a free copy of the book. Because the first five people that comment on my LinkedIn post when the episode comes out, so every episode that comes out, I post on LinkedIn that the episode is out, and so you can go to my LinkedIn page, find the post, and if you’re one of the first five people to comment on the post that you would like a copy, O’Reilly has very kindly agreed to send a digital copy to the first five people who comment. And that way we can guarantee no matter where you are in the world, we can send you the copy of it. There’s no shipping issues or… Yeah, that kind of thing.
Kate Strachnyi: 00:09:30
Do they have to say anything in that comment? What do they have to comment?
Jon Krohn: 00:09:34
“I want book.”
Kate Strachnyi: 00:09:35
“I want book.” There you go.
Jon Krohn: 00:09:40
So yeah, so thank you very much, O’Reilly, specifically. Suzanne Huston, she’s been great. We’ve had these book giveaways before and she just recently, shortly before we filmed this episode, Suzanne confirmed that she’d love to do this whenever I’ve O’Reilly guest on the show. So that’s awesome. And for those of you who don’t happen to be in those first five, but you want to read Kate’s book anyway, you can get a 30-day free trial of the O’Reilly platform with our special code that Suzanne also just gave me. It’s SDSPOD23, and maybe that’ll only work in 2023, I don’t know. If you’re listening in the far future it might not work, I don’t know. But SDSPOD23, it gets you free 30-day trial of O’Reilly, and Kate’s book is obviously in there along with millions of… I don’t know if millions, a lot.
Kate Strachnyi: 00:10:26
I don’t know if millions but thousands.
Jon Krohn: 00:10:26
Just lots of books.
Kate Strachnyi: 00:10:30
It’s in the thousands, yeah.
Jon Krohn: 00:10:32
And tons of content in there. I have dozens of videos and my book is in there. Anyway, so very cool. Thank you. Thanks to O’Reilly. I’m sure there’s lots of listeners out there that’ll take advantage of that really kind offer. So your book, it’s called ColorWise, a Data Storyteller’s Guide to the Intentional Use of Color. Let’s unpack that for our listeners. What is a data storyteller? Is it someone who makes things up about color? That sounds a lot like you.
Kate Strachnyi: 00:10:58
Right, right, right. You’re close, but not really. So a data storyteller, it can be anyone who’s telling a story and they’re using data in their story. So they’re using visuals, they’re using graphics, charts, bar charts, pie charts, whatever you might want to use to basically provide those insights in a way that’s easily digestible to their audience. So you can be a data storyteller. Anyone could be a data storyteller. All you have to do is tell stories with data.
Jon Krohn: 00:11:29
Yeah, I think we recently had an episode very much focused on a kind of storytelling aspect. So we had… Do you know Ann K. Emery?
Kate Strachnyi: 00:11:38
Yes, yes. She’s spoken at DATAcated conferences before and she did stuff. She actually teaches an Excel course in the DATAcated Circle. So familiar with Ann, yeah.
Jon Krohn: 00:11:48
What’s the DATAcated Circle?
Kate Strachnyi: 00:11:50
It is an online community for data professionals where they all get together, there’s a discussion board, and we have courses on data to dashboard with things like Excel, which Ann taught, and there’s Tableau and there’s Power BI and R and Python and all these cool things and data storytelling. So there are courses and a discussion board.
Jon Krohn: 00:12:09
Cool. Can someone get a certificate in storytelling?
Kate Strachnyi: 00:12:12
Not yet, but the platform I’m using, Circle, they’re working on becoming more of a learning management system where you can actually get certificates if you take a course.
Jon Krohn: 00:12:21
Cool. I’d love to be certified in that. People are always saying that I’m telling really bad stories. So that’s very common feedback for me. That’s why they gave me a podcast. So yeah, Ann Emery’s episode is number 637 if people want to check out a lot of guidance on storytelling in general. Yeah, so you are a storyteller too, but our specific focus in this episode is on color, of course, because of your ColorWise book. So how does that tie into the storytelling, and why does the intentional use of color matter?
Kate Strachnyi: 00:12:58
Right, so this actually came about in an interesting way where every time I would get on a podcast or do a training or workshop, I used to talk mainly about data storytelling, data visualization and visual best practices. And out of those three things, I usually would talk about selecting the right chart, making sure you reduce any unnecessary clutter. And then the third thing I always talked about was intentional use of color, because I think those were the three components that I would envision when you’re trying to put a visual together, tell a data story. Those are the three things that need to go right, right? And I personally, I just fell in love with color, because I think there’s a lot of content out there on the other pieces of what charts to pick and the clutter and just visual best practices in general, and I did not find a specific book on color for data visualization.
00:13:48
And I had two ideas right away. One, no one cares. Who cares about color? And then I’m like, “No, that can’t be it. The second one, second thing that I was thinking was that maybe no one’s thought of writing a book on this topic yet, because like yourself, they’re probably like, “Well, how much is there to say on color?” And there’s a lot, Jon. You can include over 130 pages of content according to ColorWise on the topic. And I think it ties in very well into data storytelling because color is so helpful for people to interpret information. It makes visuals pop, it makes you have emotions. It can set up alerts or highlights in your brain of like, “Oh, this is important, this is not important.” And I mean, it also does make things prettier. I know that’s sort of a last thing on there, but it can be more visually appealing, where people actually want to look at your charts.
Jon Krohn: 00:14:46
I think that even that final point there about the aesthetic is more important than people appreciate, and probably also a lot of technical people like myself and maybe a lot of our listeners, were concerned about things being correct and being sharp and crisp. But then when you are presenting information to somebody else to digest, subconsciously at least, if something is aesthetically pleasing, they’re probably more likely to accept the story that you’re telling them.
Kate Strachnyi: 00:15:16
Yes, and I have to give credit to this reference to Dustin Schimek when I was on his Data Ideas podcast. He talked about color being the last inch of data analytics, which I absolutely love that. We talk about the last mile of data analytics as being data visualization, and when he called it the last inch, I’m like, “Oh, that makes so much sense to a runner”, right? Where it’s like the last inch of the marathon, it’s the final detail. So you can make sure that you’ve got everything correct as a technical professional, you’ve got all the code right and the data is clean and makes sense and all that, but if nobody really wants to look at it, then have you wasted time? I mean, hopefully not, right? Hopefully, people will still be looking at what you’ve put together, especially if it’s important enough, but making it visually appealing and adding color to help tell that data story can really help you get that message across.
Jon Krohn: 00:16:10
Nice. Yeah, agreed.
00:16:14 As we often discuss on air with guests, deep learning is the specific technique behind nearly all of the latest artificial intelligence and machine learning capabilities. If you’ve been eager to learn exactly how deep learning works, my book Deep Learning Illustrated is the perfect place to start. Physical copies of Deep Learning Illustrated are available in seven languages but you can also access it digitally via the O’Reilly learning platform. Within O’Reilly, you’ll find not only my book but also more than 18 hours of corresponding video tutorials if video’s your preferred mode of learning. If you don’t already have access to O’Reilly via your employer or school, you can use our code SDSPOD23 to get a free 30-day trial. That’s S-D-S-P-O-D-2-3. We’ve got a link in the show notes.
00:17:04
And so I’m curious, have you always been passionate about color? Is this something that…? Because you kind of told us how you came across this idea of how there are three main areas that people need to get bright and do the visualization and color is what seemed to be the part that was least covered in the market.
Kate Strachnyi: 00:17:21
Yeah.
Jon Krohn: 00:17:21
That there wasn’t already a book out there for data visualization, but this, did it also kind of tickle your brain in a way that you were like, “Yeah, this is something I’ve always been passionate about anyway. I’d love to do this”?
Kate Strachnyi: 00:17:33
Yeah, I think it actually started when I took on my very first data visualization project when I was just getting into data visualization at work. I was given a project to visualize a scatter plot that sort of had this overlay of four quadrants. It was sort of like a personality test where you take a little assessment and then plots you on the scatter plot of four different personalities in the corporate world. And as I designed it took me forever to figure out how to do this in Tableau, how to get the scatter plot just right with the image in the back, so it was years ago also, so it wasn’t that easily done. And then the final step was I had to make it visually appealing when I was presenting it to the managers there.
00:18:15
And I wanted to use the four colors that were already associated with those personalities because there was a lot of brochures and literature on this, and personality A was red and B was green. And then the person I was reporting to, he’s like, “Oh. No, Kate. No, throw that out. We need to use the brand colors.” So I’m like, “Okay. Yes, boss. Right, whatever. I don’t care.” It’s color. I did all this work, like 99% of the work is done, I just had to just edit the colors. So I go in, I edit the colors, I make them brand colors, we present it to the managers’ managers, and they’re like, “This looks perfect, I wish we used the same colors as we used for those four personalities.” And I’m like, “Oh my God.” I knew it. I knew that we had to use those colors.
00:19:00
And I think that’s when the excitement and passion for the topic was born where I’m like, “You can make a big difference by using the different colors.” And I think that was what? Nine years ago, when that first took place, and I think from all the other visual best practices, that was always my favorite because it always surprised me when people don’t think about this. My husband got into data visualization shortly after I did, and he put together a chart, just for practice, using Power BI or Tableau, and it was a chart for Burger King, just regular Burger King calories and some other stuff.
Jon Krohn: 00:19:40
He’s a huge Burger King fan?
Kate Strachnyi: 00:19:42
No, not at all. He just found sample data. He wanted to create a dashboard so he could show other people his dashboard. So he creates this dashboard that has Burger King data on it, and I noticed that he used a lot of purple, lots of different shades of purple, and then there was green in there. And to me, I’m like, “Why in the world would you use so much purple for this dashboard that’s about Burger King?” It didn’t make any sense to me. And to him, he was like, “Well, it’s just color, it doesn’t matter.” And I think that was the second moment where I’m like, “No, it does matter and I’m going to make people care.”
Jon Krohn: 00:20:12
“I’m going to prove it. You’re going to regret saying. I’m going to write a whole book about it.”
Kate Strachnyi: 00:20:17
Seriously.
Jon Krohn: 00:20:20
“And you’re going to have to watch the kids the whole time I’m writing a book.”
Kate Strachnyi: 00:20:22
Well, they’re in school now, so it’s much easier. But yeah, exactly. Hopefully I proved them wrong. And I think people do care about color and they’re going to care more and more. When people tell me they’ve read the book, they always come back and say that they’re noticing color everywhere and they’re noticing how it’s used properly and improperly and that there is a right way to use it.
Jon Krohn: 00:20:44
Is that because you managed to find a way to print LSD into the pages of your book?
Kate Strachnyi: 00:20:49
No.
Jon Krohn: 00:20:51
People are like, “All right, I’m just seeing lots of color. I’m just noticing all the colors. Wow. It’s like a [inaudible 00:20:58] spilled out there.”
Kate Strachnyi: 00:20:58
“Dude, they’re winding together.”
Jon Krohn: 00:21:00
I’m going to read your book again soon, but not too soon because that was too intense. So yeah, can you tell us why is purple bad for Burger King? Because it’s not in their brand colors or because burgers aren’t purple?
Kate Strachnyi: 00:21:15
It’s both, right? So I either pick natural colors that you would find in a burger or use the brand colors. Exactly. So it could just tell a better story. The purple was sort of distracting. Like, “Why is this purple? And why are these numbers green?” And you can use color intentionally to tell a story, and what I’m noticing is a lot of people simply throw color on the data visualization as a last second thing, and then don’t think twice about it.
00:21:44
Even a couple days ago I posted on LinkedIn, there’s a poll, I’m asking people which colors they would use to specify different genders, right? Because right now people are sort of against using the blue for men and pink for women. And there’s this whole debate. So I just wanted to see what are people thinking. And a lot of the comments came back with, “Oh, who cares? I’ll just use any color, the first color that I see.” And I’m like, “Well, no. You could be intentional.”
Jon Krohn: 00:22:10
Right.
Kate Strachnyi: 00:22:11
It’s like when you’re getting dressed to go out, yeah, you’ve showered and brushed your teeth, but you’re going to put on to look…
Jon Krohn: 00:22:16
Yup, yup. Taking notes?
Kate Strachnyi: 00:22:22
You’re going to put on clothes that look good. You’re going to think about it. Obviously not everyone’s going to think about it, but most people, if they’re going to a special event or something, or you’re going to be presenting to a thousand people, which your data visualization might be seen by a thousand people, you would dress it accordingly.
Jon Krohn: 00:22:37
Actually, it’s funny that you mentioned that, in terms of the colors someone should wear, because my operations manager, Natalie, she recently came across a website called ColorWise that is about based on features of your skin tone and your eyes and your hair color.
Kate Strachnyi: 00:22:56
Yeah.
Jon Krohn: 00:22:56
It’s like a recommended palette of colors to wear, and she was like, “Is this Kate’s website? And I was like, “I don’t think so.”
Kate Strachnyi: 00:23:06
Oh my God, I should trademark this. I actually did. I should check out that site. See what’s happening there.
Jon Krohn: 00:23:12
Yeah, you can find out finally what colors you should have been wearing the whole time. So this is going back a little bit to something you mentioned earlier. And I don’t want to spend a huge amount of time on this because it’s not really related to color, but it’s just so fascinating to me to think about how nine years ago, when you first got interested in color, you had a boss and that boss had a boss and how happy are you now Kate Strachnyi? Not [inaudible 00:23:40].
Kate Strachnyi: 00:23:39
Oh my God. I’m so happy. I wake up every morning at 5:00 AM and I can’t wait for every single day. Monday, Tuesday, Saturday, it does not matter. I look forward to every single day. So very, very, very happy.
Jon Krohn: 00:23:53
You know Wednesday’s the day after Tuesday? But whatever.
Kate Strachnyi: 00:23:56
It is, it doesn’t matter. That’s the point.
Jon Krohn: 00:23:59
Yeah. Oh, and that also, that reminds me, I don’t know if you’re still doing this, but I remember you telling me or posting, I can’t remember, I somehow came across that you were experimenting with a reduced work week.
Kate Strachnyi: 00:24:10
Oh, yes. That was me. I still am, since June of a year and a half ago. I don’t take calls on Mondays and Fridays, and I only work between nine and two. And then even then I sort of take two hours of a break for coffee and catch up and stuff like that, and I don’t work weekends. And I take July and August almost fully off. It’s the best.
Jon Krohn: 00:24:34
Wait, wait, wait, wait, wait. So you work Tuesday through Thursday from nine to two with a two-hour break?
Kate Strachnyi: 00:24:41
Yes. But there might be times when I’m just so excited to work on something that I will work outside of those hours, but…
Jon Krohn: 00:24:49
We are filming this episode on a Friday.
Kate Strachnyi: 00:24:51
I know. And that’s why that time slot was available. I don’t have calls on Fridays, so it leaves me to do things like this. It gives me extra time to read books, go running.
Jon Krohn: 00:25:01
Right, if you’re passionate about something, you might still do it.
Kate Strachnyi: 00:25:05
Oh, yeah. I mean, I’m not just going to sit. I’m not. I want to do something, so…
Jon Krohn: 00:25:10
I better make sure that this episode is really a grind for you, that it really feels like work.
Kate Strachnyi: 00:25:15
Oh yeah, it’s so difficult.
Jon Krohn: 00:25:17
I’m going to make you redo all the colors on this part.
Kate Strachnyi: 00:25:21
I might not be here to the end.
Jon Krohn: 00:25:24
And then having you do them back to the original colors afterward.
Kate Strachnyi: 00:25:29
And the thing is, it was so quick and easy to actually make the change, but it was so reassuring knowing that I was right.
Jon Krohn: 00:25:36
All right. So I do want to spend just one more question on this whole work thing, which is, so we’ve talked about now, that story nine years ago is why you’re passionate about color, do you have any specific stories as to when you were like, “I’m working on my own. I’m going to just have my own companies now”?
Kate Strachnyi: 00:25:59
I mean, I wanted my own company since I was 10 years old. When I came to America, when I was nine, I started a company where I would print custom bookmarks for people, which is actually hilarious because yesterday I wrote down something in my notebook where I’m doing a book giveaway at a conference that’s coming up in a couple months, and my plan was to print custom bookmarks, and I’m like, “Oh my God, I’m coming full circle to my bookmark days.” Because I used to sell these bookmarks to my classmates.
00:26:29
Back then Pokémon was cool, so I would go to the local library, because we didn’t have a computer, we just moved to America, and I would print out the Pikachus and the other… Charizard or whatever, that they would request. Get their name on there, I would color them in because the library didn’t have a color printer, so I would color it in and I would sell it to them based on how many stickers or little unique things they wanted on their bookmark. Until my teacher is like, “Oh wow, you’re selling…? Oh, you’re giving out bookmarks to your classmates.” And I’m like, “Oh, I’m selling them. Do you want to buy one?” And then she’s like, “You can’t sell on school grounds” and that was the end of my business.
Jon Krohn: 00:27:07
Oh my God.
Kate Strachnyi: 00:27:08
So I always knew I wanted to do this, and I think I could have started sooner, but I’m just in such a good place right now. I have zero regrets. And I think most of the times when I realize, “Wow, this is cool” is when I see friends or family tell me like, “Oh, I just have to jiggle the mouse for another hour while I wait till five o’clock.” And I’m like, “You don’t have work to do?” And they’re like, “No, I just have to pretend I’m at work. Kind of cool.” “That sucks.”
Jon Krohn: 00:27:37
And that’s the kind of job that the educational system, that doesn’t even let you be entrepreneurial, is setting you up for.
Kate Strachnyi: 00:27:43
No, seriously and…
Jon Krohn: 00:27:44
That’s what the man wants. They’re like, “No selling on school grounds. We need you in a job where you’re jiggling your mouse for 40 hours a week.”
Kate Strachnyi: 00:27:54
Yeah, seriously. And some of these are just very high level professionals that are like, “I just have to pretend I’m working until the clock runs out.” Or, “I only have an hour for lunch.” I’m like, “Oh man, I do not miss those days.” Being in control of your own schedule is the best thing in the world.
Jon Krohn: 00:28:11
Preaching to the choir here, that’s for sure, Kate. And it’s interesting how we do kind of have this shift towards, I think, especially with… Remote work isn’t as abundant as you might think. There’s a lot of talk about remote work and a lot of people had remote work in the pandemic, but LinkedIn recently published some stats that showed that only 15% of job opportunities in the US have remote work options. But nevertheless, I think more than ever there is this understanding that forcing people to work specific hours is not the optimal way in a lot of kinds of roles.
00:28:48
So that kind of system was created at a time when we had factories and you needed to all be there at the same time because otherwise you’d need some widget to be made for you to stick that widget onto some other thing, and if the person making the widgets wasn’t there, you can’t do your job. So everyone needed to be there manufacturing the bits. But in a lot of jobs now, in this more digital service economy, you can be doing your work and slacking somebody that your part of the work is done and it can be asynchronous. So that’s kind of a cool shift that’s happening a bit, but obviously not as free.
Kate Strachnyi: 00:29:28
It’s happening slowly. It’s happening slowly, that’s for sure.
Jon Krohn: 00:29:32
Anyway, we were talking about color?
Kate Strachnyi: 00:29:34
Yeah.
Jon Krohn: 00:29:37
All right, so we asked about whether color is always necessary in data visualization. Oh no, I hadn’t asked that yet. Kate, is color always necessary in data visualization? So yeah. Are there circumstances where we don’t need it?
Kate Strachnyi: 00:29:49
Yes, absolutely. You don’t always need color, but just to keep in mind, gray, white and black are also colors. Right? So…
Jon Krohn: 00:29:57
Whoa! Really?
Kate Strachnyi: 00:30:01
I know. So we don’t always need to overload the visuals with the colors of the rainbow. I actually highly recommend that if you’re creating a data visualization, you start with a black and white or grayscale mode and then take a step back and think, “Do you understand this chart or graph without any color?” And then think through, “Who’s my audience? What am I trying to tell them?” And then see if you can properly use an intentionally planned color scheme to highlight specific data points that can help you tell that story. That’s like the whole thing. That is as simple as that, and I think people miss that probably because a lot of the tools, technology that they’re using, it comes preloaded with just so much color.
Jon Krohn: 00:30:47
Right, for sure.
Kate Strachnyi: 00:30:47
And it’s like, “Do you want these six colors or these six colors?” And you’re like, “Well, do I need color?” Start with that.
Jon Krohn: 00:30:54
That is a really good point. So, you start with assessing whether you need color at all. You can use grayscale to start and then add color intentionally one by one?
Kate Strachnyi: 00:31:05
Yes.
Jon Krohn: 00:31:06
And you probably have things like the most prominent color that you’re going to have is maybe your first choice, and then you have kind of background colors. Cool. So yeah, is that the whole thought process that one should follow when selecting a color scheme for data visualization? Or do you have some other tips?
Kate Strachnyi: 00:31:29
Well, I have a lot of tips. I have a whole book full of tips, but I’ll share a couple more here. So I think it’s important to consider, “Who is this visual intended to be seen by? So who is the actual audience? Who is the reader of this chart?” And then think through their perceptions, their cultures, their associations with color, and make sure that you use colors that won’t offend them, that won’t confuse them, that won’t make them want to not look at this chart for one reason or another.
Jon Krohn: 00:32:01
Did you say offend?
Kate Strachnyi: 00:32:03
Yes.
Jon Krohn: 00:32:03
Some colors offend?
Kate Strachnyi: 00:32:05
Are you offended right now? Yes, absolutely. So let’s say you’re talking about death and death for whatever, something really bad happened and you’re reporting on death, and you’re using colors like hot pink, bright yellow, and things that make people want to smile, it’s sort of contradictory. It doesn’t make sense. It’s almost as if the colors are conveying that you’re happy about what’s happened, and you do that unintentionally because sometimes the software is like, “Here, use this pink and yellow vibrant colors to spice it up”, where you should be using something like maybe a deep purple or black or something like that, right? To signify it’s bad.
Jon Krohn: 00:32:46
Purple, bad for burgers, good for death.
Kate Strachnyi: 00:32:48
Yeah. That should be the name of the episode. Seriously, I love it.
Jon Krohn: 00:32:56
Cool. Sorry, and I interrupted you. You were…
Kate Strachnyi: 00:32:57
Oh my God.
Jon Krohn: 00:32:59
You were talking about other tips and I completely spoke over you. It’s probably going to be impossible to remember that now.
Kate Strachnyi: 00:33:03
No, no, no. There are a lot of tips. So back to the audience, right? So not only do we need to think about what are they thinking psychologically, culturally, we also have to make sure that they can actually see color properly. So taking accessibility into account and thinking through the fact that there is a percentage of men and women that cannot see, differentiate different colors as clearly. So red and green, for example, is a pretty common color form of colorblindness where you might not see any differences between the similar shades of green and red, which is crazy because I mean, hello, the traffic lights, stock market, red and green is everywhere.
Jon Krohn: 00:33:54
Wow. Wow, I don’t know how I’d never thought of that. It’s so obvious. That’s such a really beeping glare.
Kate Strachnyi: 00:33:55
I know.
Jon Krohn: 00:33:55
Man.
Kate Strachnyi: 00:33:56
That’s why it helps when it’s red on top, then yellow, then green on the bottom, so at least positionally.
Jon Krohn: 00:34:03
Right.
Kate Strachnyi: 00:34:03
But yeah, if you were one of the unlucky few who can’t see the differences, then you’re pretty much seeing it as a mix. If you were to mix green and red together, it’s like this nasty looking color, that’s what they see.
Jon Krohn: 00:34:17
Like a brown.
Kate Strachnyi: 00:34:19
An ugly brown.
Jon Krohn: 00:34:20
Yeah.
Kate Strachnyi: 00:34:20
Not like a chocolate brown.
Jon Krohn: 00:34:23
I remember that from mixing colors together as a kid. It seems like any…
Kate Strachnyi: 00:34:26
Last week, right?
Jon Krohn: 00:34:29
My finger paints.
Kate Strachnyi: 00:34:31
Yeah.
Jon Krohn: 00:34:31
It’s my stress relief.
Kate Strachnyi: 00:34:33
Yeah.
Jon Krohn: 00:34:34
Actually, I don’t want to admit to this, but Kate, it’s actually toe painting.
Kate Strachnyi: 00:34:38
Oh my God, perfect.
Jon Krohn: 00:34:40
Yes.
Kate Strachnyi: 00:34:40
Because you broke your finger, right? That’s the…
Jon Krohn: 00:34:41
The Christmas card I sent you was toe painted.
Kate Strachnyi: 00:34:44
Very nice. I’m glad.
Jon Krohn: 00:34:45
That’s an authentic. It’s not.
Kate Strachnyi: 00:34:48
It’s on my fridge, I’m going to have to take that down, so…
Jon Krohn: 00:34:51
Okay. Yeah, those are really great tips. One of the concepts that I came across in your book, that I didn’t know about before, were the three color scales. You want to tell us about those?
Kate Strachnyi: 00:35:03
Yes. So three types of color scales that you can use. There’s categorical, there’s sequential, and then there’s diverging. So categorical you might use for data points where you’re comparing, let’s say, the four regions of a country. There’s region A, B, C, and D, and you can probably have four different colors that distinguish those regions. Something like sequential color scale is when there’s a natural progression. So for example, if you’re plotting the growth of a tree or a plant, you might use different shades of green. As the tree grows, you start with a lighter shade and it gets darker and darker and darker.
00:35:45
And then there’s diverging, and diverging color scales essentially is when you start with a low point and then you have a midpoint and then you have a high point. So generally, for divergent color scales, I’d use something starting with an orange, having a gray-ish midpoint, where you have the sort of base value, and then as the value start to go up, it gets lighter blue and then darker blue. And an example for this would be profits of a company. So if you’re in the negative, you are that orange-ish color. If you’re at your benchmark, you’re at the gray color, and then as you get more profitable, you get more blue.
Jon Krohn: 00:36:25
Right. Not green and red. Come on people.
Kate Strachnyi: 00:36:27
Not green and red. No.
Jon Krohn: 00:36:32
Cool. Yeah, especially that diverging one. I hadn’t really thought of that as a separate category. It’s interesting how there’s analogies in a way. I’m probably going to mess this up, but I’m thinking about how there’s kind of analogies to the way we think about numbers and statistics. So for example, categorical is categorical variables. How you have binary categories or multiple groups where you have a statistical or machine learning model that’s predicting whether a stock is going to go up or down. So you have these two categories that you’re trying to predict discreetly and then sequential, I mean, I think it really is the same. You have this idea of rank.
Kate Strachnyi: 00:37:21
Yeah.
Jon Krohn: 00:37:21
It can be important. You can have statistical models that predict the rank. So the distance between first and second and second and third doesn’t matter in that case, but you’re trying to get them in the right order. And then diverging, well, that doesn’t really… I guess that’s kind of like… It’s not exactly the same, but when you have a regression model, the baseline is zero where there’s no effect, and then you could have positive or negative values. Anyway, I don’t know. I don’t why I’m…
Kate Strachnyi: 00:37:52
You’re really pushing that analogy now. You’re really trying to make it work.
Jon Krohn: 00:37:57
I really was trying to make it work. But yeah, I thought there was something there. But yeah, so categorical, sequential, diverging, those are the three kinds of color scales. Super helpful to be able to think of things in those terms as we intentionally look at our charts starting from grayscale and start to add color into them. Cool. So you touched a moment ago on cultural considerations.
Kate Strachnyi: 00:38:20
Yeah.
Jon Krohn: 00:38:22
So that’s not something that I typically think about. So what kinds of special considerations do we need to have with respect to culture? I guess if you were presenting to an audience that wasn’t a culture that you’re already very familiar with, would you Google what colors mean to them?
Kate Strachnyi: 00:38:42
Yeah, you can Google. You can ChatGPT this type of information, or better yet, you can talk to somebody, right? So I’d say if you don’t know your audience very well, or if you fully expect to have these three specific cultures that you’re not familiar with, reach out to members of that audience and get a feel for, “Hey, how does this make you feel?” If you don’t have the capability to actually talk to the audience and sort of test your visualizations before you make your big presentation or send across your chart for review, then yeah, you can start by Googling. Because there’s plenty of information out there that sort of goes into, “Well, red is interpreted as something urgent or bad or maybe exciting in the Western culture” but if we look to the East, it sort of signifies happiness, luck, and prosperity. And when the stock market’s crashing, we’re coloring it red, where for them, if the stock market is rising, they’re coloring it red. So, I think that’s the number one example. It’s like, “Oh my God, it’s completely opposite and conflicting.”
Jon Krohn: 00:39:59
Yeah. That is literally the example that I was about to give is that Chinese stock markets, for example, red is a good day and green is a bad day.
Kate Strachnyi: 00:40:04
And then to those of colorblindness, you’re out of luck.
Jon Krohn: 00:40:08
I don’t know, every day is just kind of the same poopie brown.
Kate Strachnyi: 00:40:12
Yes, exactly.
Jon Krohn: 00:40:15
Cool. Yeah, so cultural considerations. Accessibility, we’ve already kind of talked about, so avoid red and green. There’s also some people that are yellow, blue, color blind, right? But that’s rarer.
Kate Strachnyi: 00:40:25
That is rarer. But, I mean, there are filters available online like Coblis Colorblind. If you just Google Colorblind filter, you can see what you look like to somebody whose color blind. You could see what your outfit looks like. You could see what your data visualization looks like. You can even install, I think, a Chrome extension where you can just see everything in the color blind filter, so you can really design for that. In addition to colorblindness, you also need to consider if whatever you’re working on and you’re spending all your efforts on designing and coloring, it might be printed in black and white, right? Or it might be viewed on a tiny screen from far away, or it might be enlarged to the point where it’s super pixelated. And planning ahead for those types of scenarios, especially the printing, I think people still print, right? If they do, a lot of times it is black and white. We have a copier in a library that I was just at, and it was all black and white. I’m like, “Okay”, so…
Jon Krohn: 00:41:27
That’s still where you’re doing your printing?
Kate Strachnyi: 00:41:30
And I have a printer, I use printers. I [inaudible 00:41:34].
Jon Krohn: 00:41:34
For your Pokémon bookmarks for your kids. Are your kids into Pokémon?
Kate Strachnyi: 00:41:40
No, but my sister’s kids are, so I’m always on the lookout for getting them stuff. But anyways, when your stuff is going to be printed, you have to make sure if you’re relying on color, to help you tell that story, then once it’s printed in black and white or it’s viewed differently on different screens, then your story’s not going to make sense anymore. And then that brings us back to starting in the grayscale. If your story already made sense in grayscale and you were able to deliver your insights, then color is really just an added benefit and helping you to really just really highlight that story. You can also use other things. You can use symbols, you can use patterns if you want to truly make sure people see the differences between two categories or something.
Jon Krohn: 00:42:26
Right, right, right. Right. Yeah, I guess in a lot of situations, if you’re making a line chart, for example, there’s no reason why you couldn’t have squares for everything that’s orange and circles for everything that’s green and so on.
Kate Strachnyi: 00:42:39
You can have a solid line and a dotted line or… Makes it real simple.
Jon Krohn: 00:42:45
Cool. Okay. So that covers cultural understanding, considerations, accessibility considerations. Is there anything else that you haven’t covered already with respect to just the general psychology of color that’s really important?
Kate Strachnyi: 00:42:58
In terms of the psychology of color, I think there are specific colors, and this brings us back to the cultural perspectives, right? Making sure that when you’re selecting the color scheme for your visualization, that it makes sense for the data. So we talked about the deaths or a funeral being considerate about what the topic is. Not only your audience, but what is the topic that you’re covering for that audience? For example, if you’re delivering this to teenagers and they might love the hot pink and the aqua blue, you still have to consider the actual content that you’re delivering to that audience and take that into consideration.
Jon Krohn: 00:43:36
Right. Cool.
Kate Strachnyi: 00:43:38
I think the last tip, which I should have probably started with, but I’ll end with, is the number of colors that you’re using. So if you have, let’s say, 15 categories and you’re trying to show sales across these 15 categories, and you’ve got a heat map going or line chart, you don’t need 15 different colors. You don’t need more than, I’d say five colors, in your entire data visualization or dashboard. More than five and people are like, “Okay, this is pretty, but what the heck is it?” Right? It might be good for grabbing attention. You’ll notice the cover of ColorWise has more than five colors, but that’s on purpose. That was intentional, and were not trying to tell anything with that book graph thing that O’Reilly came up with. It’s a chart coming out of a book with multiple colors for those who are just listening in.
Jon Krohn: 00:44:31
Yeah, it is coming out of a book.
Kate Strachnyi: 00:44:33
It is.
Jon Krohn: 00:44:35
Yeah. I didn’t know what… But the book in the background, it also kind of looks like a graph. It’s kind of like a line graph. It’s an abstraction of a book.
Kate Strachnyi: 00:44:42
Yeah, exactly.
Jon Krohn: 00:44:43
Yeah, really clever. Somebody did some really clever work there.
Kate Strachnyi: 00:44:47
Yeah, so for data visualization, you don’t need that many colors unless you’re really just trying to get attention. Many times, of those 15 categories, there’s one or two categories that actually is relevant to your specific audience that you’re talking to. Let’s say you’re talking to region A who covers categories one and two. Well, you can have two colors for those two categories and leave everything else in a light gray. So you still have those supporting details for comparison and context, but your reader is no longer overwhelmed with the amount of information that’s being thrown at them.
Jon Krohn: 00:45:24
Cool. All right. So speaking of overwhelm, so you’ve kind of covered how we can avoid overwhelming people with a single visualization, but what about considerations once we start blending lots of visualizations together? So in your book, you state that we create 2.5 quintillion bytes of data every day. I don’t even really know what a quintillion is. I can’t picture a quintillion of things in my head. Big number.
Kate Strachnyi: 00:45:50
Let’s count to one quintillion. Ready? Hold on.
Jon Krohn: 00:45:52
Okay, here we go.
Kate Strachnyi: 00:45:52
One, two… I’m done. I’m done.
Jon Krohn: 00:45:58
Good thing you didn’t write a counting book. So with all of those data being recorded, a single chart can’t possibly expect to capture all of the important things that are happening, say in a department of a business, forget like a whole division or the whole company. So even if we make simple and consistent visualizations that convey valuable insights, is there a risk of information overload if we place a whole bunch of visualizations in the same place at once? And so are there strategies, on an organizational level, or guidance that you have for us, on how we can perform data storytelling at the right level of detail at a given time?
Kate Strachnyi: 00:46:45
Yes. I think it starts with the audience and what they care about. So no matter what you’re visualizing, you’re hopefully trying to answer a question or a couple of questions for a specific audience, and that’s where you start. So let’s say your audience is a group of people who cares about the specific topic and you know the questions that they have about that specific topic. Well, that’s the beginning of your data storytelling journey. You now have answers to those questions, because we’re talking about multiple charts here. Let’s say they care about the sales, the profit, the category that’s performing the best. And because it’s the same audience, it’s like a team of management that cares about their company numbers, so you start with the questions, and then you go to your data and you provide answers. So before you even start visualizing anything, you start providing answers, and those answers are going to be your titles for each chart.
00:47:41
So for example, if sales have decreased for the last quarter, okay, that’s your title for one chart. After you’ve gone through that process, you’ve got your questions, you’ve got your answers. Those answers are now titles for each chart. Then you go on and you create those charts that support the information that visually explain the story. And then you use color on top of all that to really drive home the point. And the one thing I’ll say here is when you’re using more than one visual, let’s say more than one chart in a dashboard as an example, you have to be really careful with using colors consistently. And I see this misused almost every single time, it drives me mad. People participate in these database competitions, and then they post all of their submissions, and some of them are amazing, right? But the beginners, the number one thing I see them doing wrong is, let’s say, going back to our example of sales by category, you have three categories, and in your first chart, you’re showing sales for those categories, and you’re using red, green, and blue. Great. Now you’re showing that…
Jon Krohn: 00:48:48
I mean, they’re using red and green. That’s not great. Come on.
Kate Strachnyi: 00:48:50
Whatever. You’re using blue, you’re using three good colors, okay? And then in chart B, you’re also talking about the three categories, and now you’re talking about the expenses for those categories. And now you’re using red, green, and purple for some reason, and now people already learned that category C was supposed to be blue, and now you’re using purple. That can really throw people off and it just confuses them. So being consistent throughout the entire process is so important. Even if you introduce new data points, let’s say now we’re not talking about categories, we’re talking about regions, you still should not use the colors that you used. Let’s say if you use yellow for one of the categories, don’t use it for one of the regions, even though it’s a different data point, that same color creates a connection in people’s minds.
Jon Krohn: 00:49:39
So you must have loved my book, Deep Learning Illustrated, because we painstakingly had color consistency across the code you see.
Kate Strachnyi: 00:49:48
It’s so important, yes.
Jon Krohn: 00:49:49
In the text and in the visuals. So the variable Y, the outcome in a machine learning model, we have that blue, the exact same blue throughout all of our figures, all of our equations in the textbook and all of our code examples.
Kate Strachnyi: 00:50:07
And I think that’s brilliant. It brings me back to one more point where people are like, “Well, I’m trying to spice it up. I’m trying to keep things interesting.” And they also do that with backgrounds for data visualizations for a chart they’ll put a pink background or, I don’t know, a dark blue background, whatever and it’s because they’re trying to do something different. Well, my thinking, it goes back to reading a book, right? You don’t see publishers of books printing on slightly shades of pink and blues and purples throughout the book because the whole point here is for people to read the text, and the best way to read the text is with a white or off-white background. Same thing for data visualization, we don’t have to just get creative for the sake of spicing it up. Be intentional.
Jon Krohn: 00:50:57
Sweet. All right. Intention, that’s the word of the day.
Kate Strachnyi: 00:51:00
Yes.
Jon Krohn: 00:51:01
Today’s episode is brought to you by the letter I. So, all right. Speaking of I, let’s get into AI. Ooh, what a transition. So you have recently generated images with AI, and I mean, this is a really exciting thing that’s happening in the AI world right now. In 2022, there was an explosion in the number of models that could generate compelling graphics from a natural language prompt. So DALL·E 2 in 2022, was able to create, in a lot of circumstances, stunning, high res images of whatever you want. We had Imagen Video from Google, it creates, not as good as the static images, but it creates videos based on some prompt that you put in. And Sadie St. Lawrence and I, in the first episode of this year, episode number 641, about data science trends for 2023, one of our main predictions is that video will get better in 2023. But anyway, regardless, we were having this explosion of models that can create visuals, and yes, you recently have been using them. So yeah, you made a post on LinkedIn. You want to tell us about that? How’d it go?
Kate Strachnyi: 00:52:30
Yeah, absolutely. So I use the tool called Canva for most of my visuals, for my LinkedIn posts, for my YouTube for my background, for whatever I need to do. It’s such a brilliant tool, and when I heard that they have this text-to-image feature, I had to check it out. So I went in there, it was just a click of a button to get this app in there. And basically, you type in anything you want. You can type in text like Elephant playing the guitar with headphones on, right? And it’ll take a few seconds, but then I’ll give you four images of an elephant playing the guitar, and it’s pretty amazing. Now, the post I made was…
Jon Krohn: 00:53:10
Are you calling me fat?
Kate Strachnyi: 00:53:12
No, no. I like elephants.
Jon Krohn: 00:53:15
She’s looking at me, I’ve got a guitar in the background and headphones on.
Kate Strachnyi: 00:53:19
Hold on.
Jon Krohn: 00:53:20
And she’s like, “Imagine a really fat, bald podcast host with a guitar and headphones on.”
Kate Strachnyi: 00:53:31
That’s exactly what happened. “Holding a white pen.” Hold on.
Jon Krohn: 00:53:37
No, that’s okay. I’ll forgive you.
Kate Strachnyi: 00:53:37
Oh my God. So anyways, it’s so good at creating images. It can even create photographs. You could tell it what sort of art style you wanted to use like Van Gogh or abstract or whatever. So I recently used it to play around with creating images of a female runner in various circumstances. So I’ve got a female runner going through space, through a forest, through a high-rise building area. And I kept trying to get it to be even more creative like, “Make this brighter” and it was just so amazing the types of images it came up with. So I made a post, I think I uploaded 12 or 15 different images, and the plan for those images was to find one that I can use as the cover for a new book that I’m working on.
Jon Krohn: 00:54:24
Nice, and we are going to talk about that book in a moment as well as we’re going to talk more about your running, which you’ve now alluded to a couple of times in the show. We haven’t talked about it explicitly, you have alluded to it. But something that I wanted to ask you first, is you mentioned Canva and particularly this new text-to-image generation tool that’s built into it, are there other tools that you highly recommend to our listeners for data visualization or other data science things?
Kate Strachnyi: 00:54:53
In terms of tools for data viz, I’ve been using Tableau, Power BI and Click and R and Python, Excel. One of my favorites is actually Datawrapper. It’s so easy to use because you don’t have to install or download anything. You just go to Datawrapper and find them online. You can even copy and paste your images, so for something super quick, I tend to use that. But Canva also has the ability to create charts and graphs, and they’ve even partnered or bought, I’m not sure what the relationship is, Flourish Studio, which is even cooler because now you can have racing bar charts or flowing sankey charts, right in your Canva presentations.
Jon Krohn: 00:55:35
Oh, wow.
Kate Strachnyi: 00:55:37
Yeah, so…
Jon Krohn: 00:55:37
Did you say racing bar charts?
Kate Strachnyi: 00:55:38
Yeah.
Jon Krohn: 00:55:40
So you watched two bars in a race?
Kate Strachnyi: 00:55:43
Yes, exactly.
Jon Krohn: 00:55:43
Who’s going to be the biggest?
Kate Strachnyi: 00:55:44
Yeah.
Jon Krohn: 00:55:45
Wow, that’s super cool. Racing bar charts and what was the flowing sankey charts?
Kate Strachnyi: 00:55:51
Flowing sankey charts. You can have…
Jon Krohn: 00:55:54
What’s a sankey chart?
Kate Strachnyi: 00:55:56
Sankey chart, where you have two categories. Basically, two sides of a chart and then you’ve got… I’m trying to visualize it here for you.
Jon Krohn: 00:56:07
I know, it’s so hard for our podcast audience, they’ll be like, “Uh?” I’ll put a sankey chart. I’ll link to the sankey chart in the show notes, but it’s like, “Man, the [inaudible 00:56:17].”
Kate Strachnyi: 00:56:17
It just shows progress. It’s good to show progression.
Jon Krohn: 00:56:18
Oh, yeah.
Kate Strachnyi: 00:56:19
This company used to have bigger market share, and the market share is shrinking over time while somebody else’s market share is growing over time. It’s a good way to visualize something like that.
Jon Krohn: 00:56:27
Yeah, I just Googled it quickly and I instantly knew what these were. I see them all the time. Yeah, so you have your X axis. Your horizontal axis would typically be time.
Kate Strachnyi: 00:56:40
Yeah.
Jon Krohn: 00:56:41
Or it’s like 2020, 2021, 2022, and then you have… Yeah, it’s like a bunch of bar… It’s like a stacked column chart, but then there’s like a flow you can clearly tie. It ties the colors together nicely between different tiers.
Kate Strachnyi: 00:56:58
Yes. Colors.
Jon Krohn: 00:56:59
And you can see how maybe one category has split into other categories in a future year or merge or grown or shrunk. Yeah, sankey chart. Cool. Yeah, I’ve used these before. They’re really cool, so our company Nebula, we’re in the human resources space. And so a visualization that we’re working on right now using these sankey charts is to model, let’s say you’re looking to hire a data scientist, it’s hard to find great data scientists, so we are creating a sankey chart where you can see what kinds of jobs people tend to have before they become a data scientist. So it’s like data scientist is the middle column of the sankey chart, and then you have data analyst, university instructor, is the things on the left that merge into the data scientist career. And then also, we show you what people tend to do later. So things like machine learning engineer, data engineer, also more university instruction.
Kate Strachnyi: 00:58:00
Okay. Yeah, this is actually really interesting. Do you have that chart somewhere where we can check it out? I want to see that.
Jon Krohn: 00:58:06
Yeah. I mean, there’s no way for me to screen share in this podcasting platform, but…
Kate Strachnyi: 00:58:10
No, if there’s a link, I actually want to see what that progression looks like for more people.
Jon Krohn: 00:58:15
Yeah, it’s in R&D, but I will try to find a way to send to have an example in the show notes.
Kate Strachnyi: 00:58:23
Yay. Nice.
Jon Krohn: 00:58:28
Good idea. Yeah. Anyway, I interrupted you. So you talked about racing bar charts, flowing sankey charts, and then I went off on this sankey chart.
Kate Strachnyi: 00:58:38
Oh yeah, we were talking about tools.
Jon Krohn: 00:58:40
Yeah. But also, I think you were about to tell us about another animated Canva tool, and I started talking about sankey charts a lot.
Kate Strachnyi: 00:58:47
Yeah, that was Flourish Studio, so…
Jon Krohn: 00:58:49
Flourish Studio, yeah.
Kate Strachnyi: 00:58:50
Flourish Studio has joined forces with Canva, so you can use Flourish the way you normally would, and you can now use it within Canva as well. And you can bring in those animated charts into your Canva, if that’s your main source of presentation materials or whatever you’re generating there.
Jon Krohn: 00:59:07
Nice. Super cool tips.
Kate Strachnyi: 00:59:08
Yes.
Jon Krohn: 00:59:09
So Flourish Studio sounds amazing and Datawrapper also sounded like a really cool tool for getting going on visualizations quickly. So yes, now the running topic.
Kate Strachnyi: 00:59:24
Yeah.
Jon Krohn: 00:59:25
So, I know that you have a new book that you’re working on with running in the title, and I know that running is a big part of your life, so let’s start with the book first. What’s your running book about?
Kate Strachnyi: 00:59:39
Okay, so it was the day before Thanksgiving 2022 when O’Reilly shipped me the ColorWise books, and in that excitement, I’m like, “Oh my God, this is so cool. Okay, now I need to write a new book.” That one was officially done that day, even though it was done a couple weeks back before they actually shipped it to me, I think that’s when the realization hit. And I had this idea of, “What if I shared with people what it takes to run a media company like DATAcated, and sprinkled in stories of my actual running?” Because those two things, I’m just so excited about them. I can write about them all day, I could talk about them all day. ColorWise was an exciting topic, but I had to do a lot of research for it. I had to do a lot of validation, confirmation, make sure I’m saying the accurate things when I’m writing this book for people who might want to become an entrepreneur or start their own media company.
01:00:34
It’s things that I’m working on. So it’s super easy to write. There’s minimal research. It’s basically, “What am I working on that day?” And so I had this idea and I started writing the book live publicly instead of taking it to myself. And the idea I first had was like, “Okay, I’ll do a daily entry of what I worked on and then at the end of the year I’m going to come back and compile it and make into a book and people are going to care.” But I was reading a couple books by Austin Kleon, one was like, Show Your Work and the other one was something about being creative. And in that book he said something like, “If you’re creating something, share it as you create it.” And that gave me this idea. I’m like, “Okay, I’m going to use Substack”, which is a platform where you can write your blogs and people can subscribe to and sometimes pay a monthly fee in order to read your content.
01:01:26
So I started writing there. I made one or two posts in November, a couple in December. I think I’m at probably 40, 50 posts at this point because I’ve just been hammering them out as I get these ideas. And the goal here is to just capture what I’m working on, all the processes, the tools, the technologies, screenshots of the actual ways I do work. Screenshots of how much am I charging clients for the kind of work I do. Behind the scenes, how to get clients, how to find clients. How to have conversations with them, how to make sure they’re happy at the end of the day, as well as talking about my running adventure.
01:02:02
So two things, it’s getting me to be more focused in my work because I’m like, “Okay, I have to have something really cool to write about, so I have to do something really cool” And the same thing for my running. And in preparation for this year, because this is going to be running DATAcated for this year, I decided I’ll have these monthly adventures. So for the month of January, I’m running a 5k, a 10k, a half marathon, and something longer than a marathon, one thing every week. And I’ve got something really cool lined up for each month. Some really interesting challenge.
Jon Krohn: 01:02:37
Wow.
Kate Strachnyi: 01:02:38
The best one I’ve got, I think in June, is going to be my first ever 24 hour run where…
Jon Krohn: 01:02:44
Oh.
Kate Strachnyi: 01:02:45
During the summer solstice, I think it’s June 21st. I’m going to start running when the sun sets and then run for 24 hours, and then I’ll write about it in the book. So it gives me something to write about. It gives me a reason to run and go on these fun adventures.
Jon Krohn: 01:03:01
Super cool. I like that idea of being intentional with your adventures for the year.
Kate Strachnyi: 01:03:06
Yeah.
Jon Krohn: 01:03:06
I need some more of that in my life. It’s interesting that you said that. So you are writing it live is the way you described it on Substack, but so you’re writing summaries of what you did, because actually when you started saying that, I imagined that you would be doing live YouTube sessions.
Kate Strachnyi: 01:03:23
Oh, right. No, that’s too much pressure. You have to brush your hair and everything.
Jon Krohn: 01:03:33
So for those of our listeners who have never looked at the cover of the podcast or watched a video version, I’m very bald. I don’t know what it’s like to comb my hair. I dream about it. Anyway.
Kate Strachnyi: 01:03:46
I meant me. Anyways, the reason I went with writing is I could do this…
Jon Krohn: 01:03:51
You are really mean to me today.
Kate Strachnyi: 01:03:53
I could do it from my phone, I could do it from my laptop, I could do it from anywhere. Whereas live, I thought about using video, but then poor AI is going to have to transcribe those videos. And I think with text it pushes me to really upload the relevant images.
Jon Krohn: 01:04:07
Yeah. Yeah, yeah.
Kate Strachnyi: 01:04:09
It gets it more ready for a book at the end of the year.
Jon Krohn: 01:04:11
Totally. The reason why I thought it might be something that you’re doing is because Tina Huang, who was on in episode number 563, she’s huge on YouTube, you probably know her. So she’s huge making videos about working in data science, particularly at big tech companies, and I don’t know if she does it anymore, but at the time that I interviewed her about a year ago, so early 2022, she was most days doing livestream studying where she was following…
Kate Strachnyi: 01:04:45
I’ve seen that. I have seen it.
Jon Krohn: 01:04:45
Yeah.
Kate Strachnyi: 01:04:47
I think I’ve seen one of her videos where she was just working on something.
Jon Krohn: 01:04:50
Yeah, and you can’t see her screen. You don’t know what she’s doing.
Kate Strachnyi: 01:04:55
Yeah, you just see her working.
Jon Krohn: 01:04:57
Yeah, and it…
Kate Strachnyi: 01:04:57
It’s an accountability thing.
Jon Krohn: 01:04:58
It’s an accountability thing. And so it’s at a set time, dozens of people show up and then they follow The Pomodoro Technique. So it’s like 25 minutes on, five-minute break.
Kate Strachnyi: 01:05:06
Yeah.
Jon Krohn: 01:05:06
And on breaks, they talk about what they’re doing or chit chat. It’s pretty cool. I think…
Kate Strachnyi: 01:05:11
It is very cool. That’s partially why I’m writing the book live, as I said, is for that accountability. Because when I thought of doing this every couple days or every day journal entries, I know I’ll lose interest about 10 days in. But what happened with this is you can subscribe for free, right? But in order to fully read the journal entries, there’s typically a paywall about 20% into the article, and you’re asked to pay seven dollars for the month or $70 for the year. That’s what Substack suggested, so I’m like, “Sure.” And it’s not like seven dollars is not changing my life right now. It’s cool, but I think for me, the coolest part was when someone decided to pay the seven dollars to see my content, I was like, “What? People care?” People care enough to pay the seven dollars. I don’t think I’ve ever paid for online content.
01:06:07
If there’s a paywall, I’m leaving and I’m going to find it somewhere else for free. It is what it is. Unless it’s like my family members featured in a Wall Street Journal and like, “Okay, fine.” I’ll pay the 2.99 or 10.99, whatever it costs, to see that article. But typically, I wouldn’t. And I think what with that, in addition to accountability, it helped me understand people actually want to see this content enough to pay the seven dollars, so it encourages me to keep writing because I’m like, “They’re waiting for my content.” It gives you this, “Now I feel accountable to those people.”
Jon Krohn: 01:06:43
That’s so cool. Yeah. I love that. It’s so interesting to me. Is ironic the right word? I feel it’s one of the most incorrectly used words, but maybe there is an irony in that you never pay for online content when you run a paid content platform.
Kate Strachnyi: 01:07:05
I know. I also don’t really listen to podcasts, but I’m on here, so…
Jon Krohn: 01:07:10
I know. I know, I’m actually on the same. People are always like, “Oh, you listen to this podcast? That podcast?”
Kate Strachnyi: 01:07:14
And I’m like, “Eh…”
Jon Krohn: 01:07:15
I’m like, “I don’t have time to listen to podcasts because I’m making so many podcasts every week.”
Kate Strachnyi: 01:07:20
Yeah, exactly.
Jon Krohn: 01:07:22
And also, I don’t have a commute. I don’t like listening to things. I don’t like wearing headphones at the gym. I want to be present. Anyway. Yeah, it’s interesting how that can happen. And actually, so just a couple episodes ago, we had Tom Davenport, who’s written over 20 books, many of them bestsellers, he was in episode number 647. And Tom, when I asked him on air, I was like, “Do you have any book recommendations for us?” He was like, “I know people who either read a lot of books or write a lot of books” and he did have book recommendations. Anyway, but yeah, so there you go.
Kate Strachnyi: 01:08:05
Yeah, Tom’s awesome.
Jon Krohn: 01:08:06
I was in there. Oh, you know Tom too?
Kate Strachnyi: 01:08:08
He wrote the forward for one of my books, yeah.
Jon Krohn: 01:08:10
No way. Oh yeah.
Kate Strachnyi: 01:08:12
For The Disruptors, I clearly remember asking him, because he wrote the forward for a lot of the books for authors that I respected and I’m like, “Hey…” And I know he coined the term, “Data scientist is the sexiest role…”
Jon Krohn: 01:08:23
“Sexiest job of the 21st century.” Yeah, we talked about that in his episode. Yeah.
Kate Strachnyi: 01:08:29
And when I asked him, he was like, “Sure, I’ll do it because you spelled the word forward right.”
Jon Krohn: 01:08:34
Wow!
Kate Strachnyi: 01:08:35
I’m like, “Oh my God. I’m glad I wrote this word correctly.”
Jon Krohn: 01:08:40
That is a low bar for an author.
Kate Strachnyi: 01:08:43
No, I also had DJ Patil in that book, who he co-coined that term, with and all that.
Jon Krohn: 01:08:50
Yeah, yeah.
Kate Strachnyi: 01:08:50
So there were more connections, but I just remember that was a funny thing he said.
Jon Krohn: 01:08:54
Cool. Yeah, that is funny. It was a very funny episode. Really witty guy. Unlike you, Kate, this has just been such a boring episode.
Kate Strachnyi: 01:09:02
I know.
Jon Krohn: 01:09:03
Such a bad time.
Kate Strachnyi: 01:09:04
Slugging through this thing, I guess.
Jon Krohn: 01:09:06
Yeah, but somehow we’ve made it through. And so after all of this drudgery, I’d also like to ask you, just like I asked Tom for a book recommendation.
Kate Strachnyi: 01:09:21
So there have been many books that I’ve read, I told you before that I read over 60 books just last year, but I think whenever I’m asked what my favorite book is, it always comes down to the one that pushed me the most or made me change the way I’m doing something and I’m going to have to go with David Goggins’ Can’t Hurt Me. I’m not sure if you’ve read it or if you haven’t, I highly recommend the audiobook version. You’re against headphones, but I highly recommend that one, because he records this as a podcast, actually. So it’s like he has an official book narrator who narrates the book and then between each chapter they sort of pause and then the narrator’s like, “Goggins, man. What the heck just happened? Talk to me about this.”
Jon Krohn: 01:10:11
Oh, really?
Kate Strachnyi: 01:10:11
And then he’s just on air like, “Oh, this and this happened.” And, “I can’t believe I did that.” And it’s great because he probably wouldn’t have been the best to narrate the whole book, because he doesn’t have that professional style of reading the book, and it sort of breaks it up in a really good way. So he’s got two books out now. One is Can’t Hurt Me, one is Never Finished. I highly recommend them both. They’re not data science books, they are not self-help books, they’re not motivational books, but they will help you and motivate you nonetheless to just think through any obstacle that you have and just simply be able to overcome it because he’s overcome so much, and that’s mostly what he shares is his story of losing a lot of weight, changing his whole life around and his terrible childhood and all the stuff he’s accomplished. He’s a runner, so that helps for me, right?
Jon Krohn: 01:10:58
Mm-hmm. Also to make that connection for the listeners.
Kate Strachnyi: 01:11:03
Yes, he’s a runner and he beat a lot of records. I think the pull-up record and some other crazy stuff. And actually one of my running challenges that I’ll write about for running DATAcated in March, will be the David Goggins challenge where you run for four miles every four hours for 48 hours. I’ve done that challenge.
Jon Krohn: 01:11:22
You’ve done that before, right? Yeah.
Kate Strachnyi: 01:11:23
A couple times already. I try not to miss a year ever since I’ve learned about the challenge.
Jon Krohn: 01:11:27
Ugh.
Kate Strachnyi: 01:11:28
It’s so much fun.
Jon Krohn: 01:11:29
Oh, that reminds me that you did a CrossFit Hero WOD in 2022.
Kate Strachnyi: 01:11:36
Oh, wait. It was like a… Wait, what is that called?
Jon Krohn: 01:11:39
It was on Memorial Day, right?
Kate Strachnyi: 01:11:41
Yes. There’s the Murph Challenge.
Jon Krohn: 01:11:41
Did you do the Memorial Day Murph?
Kate Strachnyi: 01:11:44
Murph. Murph Challenge, yes. That almost killed me. I did most of the stuff without the weighted vest, but that last mile I did with the… And the last set I did with the weighted vest. It was hard. I felt pain during and after.
Jon Krohn: 01:11:59
Yeah. For our listeners, yes, you’re supposed to wear a weighted vest. I think it’s 20 pounds for men, 16 pounds for women, I think it’s like the [inaudible 01:12:08].
Kate Strachnyi: 01:12:08
You use 12 pounds. They did the… Yeah. That’s the one I had.
Jon Krohn: 01:12:11
I mean, it’s serious. It’s crazy. I can’t do it Rx at the prescribed weight. I mean, not even close.
Kate Strachnyi: 01:12:17
Yeah.
Jon Krohn: 01:12:17
I don’t even know if I could do it just with my body weight because you start with a mile run with the vest on, and then it’s a hundred pull-ups, 200 pushups and 300 air squats, and then you run another mile.
Kate Strachnyi: 01:12:30
Yes. Well the running part was clearly my favorite. I’m like, “Oh, I could do that again.” Yes.
Jon Krohn: 01:12:38
Cool, Kate. Well, it’s been so much fun catching up with you to be honest about how I really did feel about this episode. Obviously, great.
Kate Strachnyi: 01:12:47
Finally. Are you being ironic right now?
Jon Krohn: 01:12:51
I just loved this episode so much. I wish it could go on forever, but sadly…
Kate Strachnyi: 01:13:00
Oh, no. Look, we’re out of time.
Jon Krohn: 01:13:01
We’re at the end.
Kate Strachnyi: 01:13:04
Yeah. No, this is definitely fun. I’m glad to have the honor to be back on the show a third time. I truly feel blessed to even have the opportunity, so thank you.
Jon Krohn: 01:13:12
No, we didn’t have to re-record anything, so favorite episode ever? Probably.
Kate Strachnyi: 01:13:18
I mean, sure for this week. Absolutely, Jon.
Jon Krohn: 01:13:23
And yeah, thanks for making it happen on a Friday. Enjoy your super long weekend.
Kate Strachnyi: 01:13:29
Thank you. Thank you.
Jon Krohn: 01:13:32
All right, and yeah, listeners, I hope you have a great week. I realize this episode doesn’t come out on Fridays, so that’s a bit of a weird transition, but maybe by chance you’re listening to it on a Friday. Hopefully I got lucky.
Kate Strachnyi: 01:13:42
Yes.
Jon Krohn: 01:13:43
All right. Thanks so much, Kate, and we’ll have to catch up with you again soon. It seems inevitable.
Kate Strachnyi: 01:13:49
Okay, sounds good. Thanks, Jon.
Jon Krohn: 01:13:56
Well, as usual, my conversation with Kate descended into some serious silliness, but she did also fit in a lot of highly educational content. In today’s episode, Kate filled us in on how the intentional use of color is essential for making data visualizations easy to understand, therefore making it easier for you to convince your audience of your data-driven hypothesis. She talked about how an effective thought process for intentional color use is to start with grayscale and then methodically add colors in one by one, up to a maximum of up to five colors. She talked about how there are three types of color scales we might like to consider depending on the data we’re working with. That is categorical, sequential and diverging scales.
01:14:34
She talked about how colors can convey dramatically different meaning depending on culture, such as green being positive and red being negative in the west, while those meaning the opposite in the east. She talked about how when using multiple visualizations in a given document presentation or dashboard, the colors of related concepts across those visualizations should be consistent and how she loves Flourish Studio for animating visuals and Datawrapper for creating visualizations lickity split.
01:15:02
As always, you can get all the show notes including the transcript for this episode, the video recording, any materials mentioned on the show, the URLs for Kate’s social media profiles, including her massively popular LinkedIn profile at SuperDataScience.com/651, that’s SuperDataScience.com/651.
01:15:21
Beyond social media, if you’d like to engage with me coming up on March 1st, I’ll be hosting a virtual conference on natural language processing with large language models like BERT and the GPT series architectures. It’ll be interactive, practical, and it’ll feature some of the most influential scientists and instructors in the large natural language model space as speakers. It’ll be live in the O’Reilly platform, which many employers and universities provide access to, but if you don’t already, you can grab a free 30-day trial to O’Reilly using our special code SDSPOD23. We’ve got a link to that code ready for you in the show notes.
01:15:58
Thanks to my colleagues at Nebula for supporting me while I create content like this SuperDataScience episode for you. And thanks of course to Ivana, Mario, Natalie, Serg, Sylvia, Zara, and Kirill on the SuperDataScience team for producing another colorful episode for us today. For enabling that super team to create this free podcast for you, we are deeply grateful to our sponsors, whom I’ve hand selected as partners, because I expect their products to be genuinely of interest to you. Please consider supporting this free show by checking out our sponsors’ links, which you can find in the show notes. And if you yourself are interested in sponsoring an episode, you can get the details on how by making your way to JonKrohn.com/podcast.
01:16:34
Last but not least, thanks to you for listening. We wouldn’t be here at all without you. So until next time my friend, keep on rocking it out there and I’m looking forward to enjoying another round of the SuperDataScience Podcast with you very soon.
