Google Data Studio Dashboard Do’s and Don’ts with Michele Kiss

[00:00:00] LP: Hello, hello. Lea Pica here. Today's guest helps data practitioners create Google Data Studio dashboards that create raving stakeholder and client fans. Stay tuned to find out who's dropping the knowledge on the Present Beyond Measure Show, Episode 83.

[00:00:16] ANNOUNCER: Welcome to the Present Beyond Measure Show, a podcast at the intersection of analytics, data visualization and presentation awesomeness. You'll learn the best tips, tools and techniques for creating analytics visualizations and presentations that inspire data-driven decisions and move you forward. If you're ready to get your insights understood and acted upon, you're in the right place. Now your host, Lea Pica.

[OVERVIEW]

[00:00:45] LP: Hey, guys, and welcome to the 83rd episode of the Present Beyond Measure Show. The only podcast at the intersection of presentation, data visualization, storytelling, and analytics. This is the place to be if you're ready to make maximum impact and create credibility through your thoughtfully presented insights and ideas.

Today's interview is loaded with juicy and actionable Google Data Studio dashboard tips from a true GDS pro, so be sure to stay tuned in. But before we get rolling, I just have a few fun updates for you.

All right, now, as usual, I am stoked for today's guest. But in particular, I've known this analytics rock star since I began my journey as a data storytelling advocate and trainer. Gosh, 10 years ago now, and she has a well-earned reputation as a total pro on building Google Data Studio dashboards. Let's dive in.

[INTERVIEW]

[00:01:52] LP: All right, hello, and welcome everyone today. I can't wait to introduce my guest. My guest is a recognized Digital Analytics leader with expertise across the web, mobile, and marketing analytics. She's a senior partner at Analytics Demystified, where she works with clients on analysis, training, and process to draw insight from their digital data. She was the winner of the Digital Analytics Association Rising Star and Practitioner of the Year awards. She's a frequent blogger, writer, podcast contributor, she speaks all over the world. I've gotten the privilege of speaking with her at many conferences, and we go really far back. I'm so thrilled to finally, finally have her on the show today. So please, help me welcome the latest guest in my superstar women and analytics spotlight, Michele Kiss. Hello.

[00:02:43] MK: Hi, thank you so much for having me.

[00:02:46] LP: Of course, this is way overdue. You were really there at the outset of my journey as I started to break into the conference circuit and get to know the very tight knit family that is in the digital analytics practitioner space. I've just always really appreciated, always admired the clarity and the value that you give in your talks, and just really appreciated all the support and friendship that we've had along the way.

[00:03:14] MK: I still remember early conferences, I remember when we first met and normally people come at least – I know that when I first started speaking, it was a bit of a train wreck up there, and then you would get up there and you're like, “Man, you nailed it.” You were brand new to me in the industry. I remember just being like, “It is awesome that we have somebody out there speaking very clearly, very well, and women and analytics, go!” It was pretty exciting.

[00:03:43] LP: I appreciate that so much. I think I remember things a little bit differently. I don't remember anything [inaudible 00:03:49]. I do remember avid standing ovations for you, including myself, because I think I called you the OG before. For me you are one of the OGs of that space, especially as a woman coming in and really like having the deep expertise as a speaker, but also, I loved how much common ground we shared in the actual communication and visualization, storytelling part of data, as well.

So, we've had so many alignments there. That's why I'm so happy to have you on the show today.

[00:04:21] MK: Yeah, happy to be here.

[00:04:23] LP: Cool. So, I mean, you're super well known in the space, but just in case someone has been living in a cave the last decade, why don't you let us know what it is you do and also your origin story. Everyone loves that. How did you fall into this crazy line of work?

[00:04:39] MK: I definitely count as one of the people who fell into it. I moved from Australia to the US in 2005, and I just had to get a job, any job after college. I ended up being employed as a marketing assistant at Kelley Blue Book, like the car valuation company. They had this interesting VP there, and he would hire people, like he didn't really want an assistant, he would just hire people, like, “I think they're smart. I think they'd fit into the company.” He would kind of hire them in, and then figure out where in his team somebody needed help.

We had lots of people that came in through this role, and went into user experience, went into editorial, and in my case, went into analytics. So, I started as this assistant, and the analytics team needed help, and they're like, “Okay, we'll show you what buttons to press in Excel.” I ended up learning everything on the job. We did a lot of advertising analytics, and a lot of – back then it was Omniture, I learned everything from a mentor that I had at Kelley Blue Book, who kind of took me under her wing. And then, the rest is kind of history.

I, over time, became the manager of the team. And then, I hit a certain point, probably, I think it was about five years in where I was like, “All right, I think I need to be challenged in different ways.” I left Kelley Blue Book, and I joined an agency figuring that gave me a lot of experience with lots of different business models and analytics challenges without having to like job hop constantly. So, I joined there, then I went back practitioner side, and then Eric Peterson from Demystified contacted me, and then that is where I've been ever since.

[00:06:34] LP: I know. Actually, I had the pleasure of working as a consultant, like a gun-for-hire, for Analytics Demystified, too. So, it was great to sort of be close during that time period, and work with one of the most stellar teams of all time in the space.

So, what I would gather from what you're saying is that you present data as part of your work. You don't just explore and crunch the numbers and make pivot tables all day, do you?

[00:07:01] MK: Definitely not. I think that would be kind of boring. So, trying to make sense of it for other people, and to make it seem as uncomplicated and easy as possible for others is kind of a big part of the job.

[00:07:18] LP: So, when did you actually discover the explaining part of data? Because I know a lot of practitioners find the explaining part to be sort of the unfortunate byproduct that they didn't realize they'd have to do if they joined as an analyst. I know it took me by surprise. And it's a very cross disciplinary set of skills. So, what drew you so much to that part of it, and also speaking? And what do you think are the qualities or skills that people should think about if they really want to excel in that area?

[00:07:52] MK: I think, definitely my own explaining and presentation skills developed over time. I can look back at work samples and things from things I did early on in my career. I'm like, “Oh, that's adorable. That was sweet that you did that. Cool.” But I actually had a really interesting role several years ago, pre-Demystified where the team that I worked on had both, at the time, we were web analysts and we also had a team of statisticians. The statisticians were doing very deep number crunching. They were the ones sitting in SAAS, and creating statistical models and things. And I would work with one of our statisticians in particular, and she was, hella smart.

But we would get into a meeting with stakeholders, and she wasn't great at necessarily being able to explain it in a simple way. So, we paired up, and I became almost like the business translator. She would do the analysis and then I would be the person that explained it to the business and was like the conduit between some of the advanced analysis work that she was doing, and how to explain this to our VP of sales, or how to explain it to our CEO. It was a really valuable role to be in, to have to figure out how to be concise, and how to explain things.

It's not about dumbing it down. It's just about making it so that you don't need an advanced degree in mathematics to be able to understand it. It was great experience, and I think it helped me later on to do the same thing with my own work because it is hard when you have been doing, like I still do hands on analysis, and you've been so like deep in something and being able to come 10,000 feet up and be able to explain it really succinctly is difficult. So, I think it was great practice for having to do that for the rest of my career.

[00:09:57] LP: It's so fascinating to hear about the evolution of things. You make a really good point that a lot of times people will come to my data storytelling workshops and say, “Are we doing Python and any kind of programming languages? Tableau?” I’m like, “No, actually. You're going to be drawing less on your experience with stats and code and more on your experience from parenting if you have kids. “Can you make a bedtime story exciting and make it really simplistic and use things like suspense and an animated way of speaking?” I even suggest sometimes to people think about how you read a bedtime story to your kids, or get a book and practice that, and now you have more of the energy type and the way of speaking, and the simplicity that you need, as you said, to get people to understand things.

[00:10:58] MK: It's funny you say that, because part of my son, I have a six-year-old, part of his bedtime routine is a made-up story, which is always something like –

[00:11:05] LP: It's a great idea.

[00:11:06] MK: He opens the letterbox, and in it, he discovers a magic portal that goes to a mystery world full of talking gummy bears or something. It's always really outlandish and crazy. But it's like part of our routine every single night.

[00:11:19] LP: I am so borrowing that idea for mine. We're always going off of pre-written things but I'd much rather the creative process, and I like how ludicrous that can be. So, that's really great. Also, I think it helps instill a sense of, you know, when I think about a bedtime story, I look at how little text is on each page. They're super visual. They'll start a sentence and then have a dot, dot, dot, and you have to turn the page. That's kind of how I try to create slides in a presentation. Are they pages that you can turn in a story? Have you created a page turner? So, I love that idea, and I hear that you resonate.

Well, one of the reasons why I'm glad you're here is people love the tools. There's one tool in particular that you are super proficient with that I can't wait to talk about. I haven't had anyone on to really talk about it, and I'm going to guess you know what that tool is. Drop it like it's hot.

[00:12:18] MK: Data Studio.

[00:12:20] LP: Google Data Studio. Yeah. So, we're going to talk about that here. I'm so excited. So first, why Google Data Studio? Why that versus all of the other data visualization options that are available out there.

[00:12:37] MK: I work with a lot of clients that live really heavily in the Google ecosystem. So, I just find that it just fits very naturally and well if you have teams that are using a lot of Google Analytics and other marketing products. Also, if you live in the G-Suite world of Gmail for work, and all of that kind of stuff, it does fit very naturally. I've used Tableau and some of the other data visualization platforms, but I like that Data Studio is just kind of there and so easily accessible for people.

I used to, when I was a very young analyst, I remember sitting in like these conference presentations, and everybody would talk about these amazing things that they did. But they did it by having expensive MicroStrategy licenses, or huge enterprises like Tableau work. I would just kind of sit there, like, “I don't have any budget and I can't do those things.” So, anything that I can immediately use to add value, and to start working right away is key for me. It's like one of the things that makes something have value in my world. So, Data Studio has definitely been one of them when it just so easily integrates with the rest of the Google ecosystem.

[00:13:56] LP: For you, time to market, and ease of implementation are really key. I will admit, for the same reasons I've used Data Studio in the past and what I was particularly impressed with, versus a lot of the other tools that we've just grown to adopt everywhere that are available everywhere, is I have a chart detox process that I take students through and it had preemptively eliminated a whole bunch of those steps before even getting started.

So, I was like, “Wow, someone was listening to The Wall Street Journal guide to information graphics.” They definitely implemented, it seems like they implemented a lot of best practices out of the gate so that they're eliminating some of the guesswork that people have, or default formatting that ends up really distracting people. So, I appreciated that as well.

[00:14:52] MK: Yeah, and don't get me wrong. There's default formatting in Data Studio that I turn off immediately. I have my process where I go through and I change things. No tool has perfect defaults but I do think that it gives you a lot of flexibility to be able to really almost start from this blank canvas and design something exactly how you picture it working, and in a way that works for people's brains. As opposed to thinking about tools, like if you think about in Adobe, and you're creating something in Analysis Workspace, and everything is boxes, and everything has very defined formatting. I like the freeness of Data Studio and having this canvas where I can control how it all appears.

[00:15:39] LP: Yeah, there is something to having a greater degree of creative license to create the shapes and the dimensions of things. Sometimes other platforms can feel constrained in terms of having two modules side by side, and then the whole page has to look that way and follow that format. So, I hear that. Obviously, this is a data storytelling podcast. I love to hear stories of use cases that people have, especially for using these tools. So, do you have any interesting particular cases that you implemented in Data Studio and saw really great results for communicating?

[00:16:20] MK: I mean, I use Data Studio for a lot of different things. I use it for building dashboards, I use it for building bigger, more complicated reports with more information than what I would consider a dashboard to have. I actually used it myself to do analysis, because I find it so easy to build something that lets me look at a lot of different data at the same time and then add a couple dropdowns, and then I can filter and I can like actually engage with the data without having to pull up 17 different tabs in Google Analytics, for example. I can really play with it. I can view like, “Oh, what does this trend look like? What does this trend look like?”

So, I actually use it as an analysis platform as well and I do build things that I don't expect anybody else will ever look at. They're just for me to process the data.

[00:17:14] LP: Right. Of course.

[00:17:17] MK: But I also like, I have some clients who I effectively build presentations in Data Studio. So, what I'll do is I make Data Studio, going back to constraints and how free Data Studio is, I will make it look like slides. I'll have my title that is talking about what the data is saying, but I use it to like live pull the data in. It means that, especially if it's like – sometimes if it's a smaller client, where maybe they don't have necessarily a ton of analytics resources, you can build something that's very easy to go back to in three months, or six months, or a year, and basically, it'll repoll all the new data. You just need to change the timeframes and then you can go through and kind of update your commentary or change the titles of the slides so that they're reflecting like what the new data says. I just kind of make it look like a presentation with dynamic elements, and I find it really great for that.

[00:18:17] LP: I was going to say, because I've always struggled a bit with feeling like, I don't want a presentation to feel super cookie cutter in terms of a slide. Like, “This is our conversions, and that's the story.” Right? I use storified slide titles, where it's like an observation of the data. But I love that you're creating a sort of format or template that the information can be updated on the fly, but then you're making small adjustments. I think that's fantastic.

[00:18:45] MK: Yeah, it's definitely a time saver, too. So, versus having to build a whole slide deck from scratch in a couple months if you decide to revisit something. It makes an analyst's life a lot easier to be able to do that and have it just update all the charts. When you do that, there's always other directions you're going to go in where you're like, “Oh, I just noticed something. Now, I need to insert a new slide with some new charts that's going to break down this observation that was in the last couple of months that wasn't there a year ago.” But just to be able to pull that data in a pretty automated way, eases a lot of the heavy lifting, and is a pretty good way for people to communicate.

[00:19:26] LP: And keep the charts looking unified and consistent and clean. Wow. So, are you using one page for each slide? Is that the general idea? Or how are you structuring that?

[00:19:39] MK: Yes, so each, like what would be a slide is one Data Studio page, and I tend to follow the, you know, you have one point you're making on each of those slides. So, often some of these decks can be long, because there might be a lot of things that I'm trying to cover in that time. Most of the time, it is typically like only one, at most, two charts per one of these. It's not like a ton of information crammed on, but it saved me lots of time in revisiting different analyses, because inevitably, somebody always is like, “Can you update that and see if that's changed?” It makes life a little easier.

[00:20:18] LP: I really want to make sure everyone heard what Michele just said, that you might end up with a larger slide deck, but that you have simpler slides with one main idea on each of them. This is one of the philosophies that I think we both got aligned on, most likely from Nancy Duarte’s perspective on the single idea per slide tenet. Even just this week, during a workshop, I was asked, it happens every time, “But I'm going to have a lot of slides.” I know my take on having a lot of slides that you use as many as you need to communicate your idea clearly and that people understand it. I'd love to hear your take on that as well.

[00:21:02] MK: We have had in previous events and things like that, my company Demystified, in the past, we've done a conference called Accelerate. At times, I've shared my deck for us to be using with my boss, and he's like, “There is no way you are getting through 217 slides in a 20-minute presentation.” I'm like, “Oh, watch me.” Because each one has such a small amount of information that it fits this flow where I'm talking and I'm flipping slides as I go. Sometimes the slide is just an image, that's the only thing that's on there, or it's one chart, or it's a specific point that I'm trying to make. But I am definitely within my organization. I am the crazy number of slides person because I don't like jamming six things onto one slide. I don't feel like when you present, you should be like, “Okay, now in the top left corner, this chart is saying that’s now in the top right.”

[00:22:02] LP: Like a zoo tour guide.

[00:22:04] MK: Yes.

[00:22:06] LP: I so hear that. And I think Accelerate was the second or third conference I ever spoke at. I got called out on the slide count, as well. There was a lot of doubt there, too. So, I think we were staging a mutiny during that one. But I think, to your point, not to divert from Data Studio but in general, I think this is a really interesting cultural trope that we're struggling with in the field, where no one really learned how to bring proper storytelling techniques to the data communication and presentation process.

The way that I think of transitioning between slides is editing cuts. When you're watching a movie or a TV show, if the frame sat in the same spot, but shoved a whole bunch of things happening in that frame, you'd eventually lose attention. Like, that's not what's actually going to keep yourself engaged. So, not a lot of people know that the general best practice out in the world is don't sit on a slide for more than a minute. I think the Duarte average, I think Nancy Duarte mentioned that their average is three per minute, which is pretty fast.

[00:23:20] MK: That makes perfect sense to me. It’s the style that I tend to follow as well.

[00:23:24] LP: I remember. I was like, “Oh, she knows. She knows.”

First of all, I love that you mentioned using Data Studio for analysis. I haven't thought about it that way as much. I've always sort of imported into Tableau and sort of messed around with stuff in there. But that's interesting to hear that you're using Data Studio and filters for analysis. But what I'd love to know is, how do you differentiate between creating a Data Studio dashboard versus a report?

[00:23:54] MK: I tend to go by the rule of dashboards as being everything in one view. So, I follow the rule that once I have multiple pages, and, “Oh, this part is talking about our traffic, and this part is talking about our conversion. And this is talking about what products people buy”, then you're talking a report. I cringe a little inside when we start calling everything a dashboard, because I'm like, “It's not.” It should be your at-a-glance operational metrics type of thing.

So, it’s semantic. I totally understand I'm making a semantic distinction between the two, but that's how I tend to think of them. When you're building something that's kind of at a glance, that's your dashboard. Soon as you start having a ton more information, you're building out a broader report.

[00:24:41] LP: I really appreciate that. I so agree, when you say at a glance, that's what really resonates for me where I keep trying to think like, what is my car dashboard for? What is the thing that this was based on? How is that build? What was that designed to help me do? Which was be able to instantly gauge the vital systems of my car and make simple decisions between doing something myself or taking it to an expert in a glance, without being a car mechanic. So, I appreciate that distinction. I want to make sure people really hear that because I absolutely have encountered dashboards that are really reports, or something much more long and involved. When do you see Data Studio being used well? What are the hallmarks of a well-used Data Studio?

[00:25:34] MK: I think the number one thing is that putting aside any tool that you use, it's used well when you are following general data visualization best practices. So, you can use an abuse any solution, it doesn't matter what it is. I've definitely seen Data Studio used poorly for those reasons. Anything that's built in a very – following the rules of data visualization, following the rules of what a dashboard is, what a report is, in a user intuitive way, kind of like a good user experience, is something that I would consider a good use of Data Studio.

But I also think that it does open up the ability to be something else, which is not necessarily just building static reports. But Data Studio allows you to create almost a safe space for other people to do analysis. So, you can do that by building the foundation of what the report looks like, but allowing user engagement with it via filters and having different dropdowns, or the ability to like change timeframes, or pick a particular like, “I just want to look at this for mobile.” For somebody who may not know how to do all of that in Google Analytics, or whatever analytic solution you're using, you can provide this like space where they can do a little data exploration, and you're not trying to build out every chart that could possibly be built on every cut of every dimension.

But you're giving them some ways that they can break things down so that they can look at like, “I want to view this by my particular country, or I want to view this by mobile or desktop, or I want to view this based on my specific landing page,” whatever that may be. But I find that that's a good valuable use of something. Data Studio is actually – it's not really built to be printed out or PDFd and build static reports. It's intended to be interactive.

So, I think successful uses of it are when that interactivity is built in in an intelligent way by analysts who know the data, know what they can do with the data, and give business users the tools to be able to engage with the data the right way. I find that that's a very helpful use of it.

[00:28:05] LP: I really liked that. I pulled out intelligent interactivity from that, and what I appreciate about the role of data practitioner is not always just spoon feeding the insights. I do think there's a place for really taking the time to craft a full story and narrate it during presentations. But also enabling and getting people excited and drawn into the process of exploring their own data, but creating a very clean sandbox for them to play on. They're not the ones trying to figure out the data hygiene and tripping up on misinterpreting something, because something wasn't supposed to be there, right?

[00:28:49] MK: Yes. I mean, everybody can't be an analyst, and so you have to have people that have different roles in your organization. Having that analyst data subset, almost, is really useful. I also find that it's valuable when it allows you to pull together other information. So, it might be pulling together multiple data sources so that somebody doesn't have to go to six different places to get a view. But it can also be integrating with business knowledge. So, adding things into Data Studio like time-based annotations that are going to give, when the business user is looking at something, it's going to give them some context that an analyst might have been managing the annotations to say, “Hey, this happened on this day. You should be aware of it as you're looking at this data.” But I think that as you can bring all of that information together, you can make something that's built in Data Studio really valuable for your business users.

[00:29:54] LP: Yes, I love the idea of creating something that is really empowering and alerting the audience as much as possible because I know on the client or stakeholder side, they can often feel like they're in a black box, and relying just on the word or however it's being communicated. It’s definitely not that they want to know all of the nitty gritty. They have their own work and priorities and sort of business disciplines that they have to worry about. But it is really great when they want to get involved and have that sort of hand holding and direction that they need. I love it.

[00:30:32] MK: Yes, and provide some level of self-service. Because you don't want to have everything that somebody wants to know always having to come through an analytics team and have a personal look at it. So, if there are things that you can do, to make some information available, like get stakeholders fingertips and give them a little bit of control, and the ability to play around with it and do some of that slicing and dicing themselves, I think that that ultimately can also free up your analysts to do more valuable work than just constantly pulling data based on somebody's win this week.

[00:31:07] LP: Being the gopher?

[00:31:10] MK: Yes.

[00:31:11] LP: Yes. Teaching them to fish. You're able to catch bigger fish when you're having more resources. Are there charts or anything that you particularly love in there? Are you pretty basic? What do you tend to lean towards?

[00:31:25] MK: I combine charts a lot in the sense that I use a lot of scorecards with line charts, for example. So, that you've got one thing that tells you the number, and one thing that shows you the trend. I use a lot of combos like that. I also use a lot of in-table bars, as opposed to a bar or column chart, because I feel like they take up less real estate and kind of give you some of that same visual at the same time.

[00:31:53] LP: The data bars, yes.

[00:31:54] MK: Yes. So, I would say that I'm a pretty heavy user of those. I also like the ability in Data Studio, where you can make charts interactive, so that if you click one chart, it's going to filter the other things in either a group of charts or on the page so that you can create dropdowns and things for people to select, but they can also literally engage with the charts themselves and have that feed through to other visualizations.

The other thing I would definitely say that I'm a very heavy user of is I build a lot of funnel visualizations, like a lot. That seems to be a lot of what people want to see with a lot of the projects that I'm working on. It's like how are users flowing down through this particular path? So, I have taken different approaches. I will sometimes use bar charts as a funnel, so everything's left aligned, but it's just kind of stepping down, and I use scorecards to create the metrics and the percentage drop offs and things.

I've also used some of the community visualizations. Some of the funnel visualizations, some of the community ones are incredibly ugly. And then there's a couple of them that are pretty good. If you tinker with them, you can get them to look pretty good and they do a lot of the math for you. So, I both kind of DIY it using base visualizations that are available in Data Studio. But also, sometimes I'll use some of the community visualizations for places where it might save me a little bit of time.

[00:33:30] LP: Interesting. So yeah, funnels are always interesting because they're sort of a lightning rod of controversy of the understanding that first, it's not really technically a funnel, it's more of a sieve. Also, the visualization tends not to be the most easily interpreted, accurate representation of both people remaining and the fall off because of that, sort of like pyramid structure. So, I've generally always used bars creatively for that. But tell me about community visualizations, because I know there are a lot of templates available out there that you can adopt, but are these individual visualization types that you can bring into your report?

[00:34:11] MK: Yeah, and you can say like, “Oh, I want a sunburst visualization or something like that.”

[00:34:15] LP: Oh, cool.

[00:34:16] MK: I would say I'm not a heavy user of them. I tend to use the default visualizations in Data Studio because I very much subscribe to the rule of, I shouldn't have to explain a chart to you. The chart should make sense. Somebody looking at it should know how to interpret it. And when I have to explain, “This is how the chart works,” it's too fancy, and it's just adding too much mental overhead for people.

So, I will confess that I don't use a ton. I tend to use default chart types much more than community visualizations, but I have had them come in handy a couple of times. They definitely do require though, if the community visualization, the person making it, at all of a sudden doesn't keep it up to date, or pulls it down or something like that, you've got broken dashboards. So, there's a little bit of risk inherent in that. There's also been times where like community visualizations broke in Data Studio. So, default charts would work fine nut the community ones didn't work for some period of time, just like random bugs with that in particular.

I think that if it's a well thought out chart that is going to make intuitive sense to people, and it exists in a community visualization, then by all means, use it. But I don't love using them because they look fancy, or just because I have too many line charts on this page. So, I'm just going to make this another chart for no reason when the data is actually suited to a particular type of default visualization.

[00:35:55] LP: You say you apologize because you stick with the basics, but I actually think that's the answer I'd be in alignment with as well. Because for the most part, the most default basic chart types are the most universally understood. They don't have a learning curve. As soon as you add the need to learn something new, you are slowing down the process of the actual communication. And that's especially problematic for a self-driven, self-consuming environment like a dashboard or a report.

What I will say is that I have a few cases where there are chart types that are not universally familiar, like a bar or like a dumbbell dot plot. But it's actually a very well-designed chart that accomplishes something very specific, which is allowing you to visually compare the difference between two values across a list of categories and also across the categories. I'd have to put up an example to show it.

I used it once in a presentation, but what I made sure to do is only reveal – I actually used the unfamiliarity of it to sort of create some anticipation, but then only revealed each part of the chart after I had explained each component, and that actually built the story as I went along. So, the big final reveal was sort of the punchline to the story of the places we really needed to focus on, that this chart was showing. But also, once they got through the explanation, step by step with me walking through, I was amazed how quickly they got it. But I know if I had shown it all at once, and said, “This is what the chart is saying.” They would have been zooming all around going like, “What is this? I don't know what this is. How do I read it?”

[00:37:43] MK: Yeah. I will use very lame methods, whether it's in slides or whatnot, to literally just – I might have the same thing on five slides, but I've just blocked pieces of it, just covered them in white, and I will reveal them slowly. Or even if people are going through the slides themselves, when they progress in that way, it definitely helps explain little pieces of information at a time so that then at the end, the whole makes sense, rather than, “Nam, here's the entire complicated thing.” And they're like, left with heads’ spinning.

[00:38:20] LP: Yeah, it's like trying to watch a Christopher Nolan movie where they explained the entire plot in one sentence. You're not going to get it right away. You have to have it paced out. That's actually why I call that technique shape pacing, where I use shapes and animate them in and out in order to reveal or obscure pieces. I always say, “Think about how Hollywood is showing you things. And then think about how differently we're doing it and ask yourself, why are people so confused?”

All right, so we talked about some of the best practices. Where do you see Data Studio not being used so well? What are some of the missteps that you see?

[00:39:01] MK: I mean, the general bucket of things that don't obey data visualization best practices. I remember being at a GA Partner Summit early on, and they were showing a Data Studio report and just the amount of crap on it was insane. I think there was like a giant picture of a lion in the background, and I was like, “What is going on?”

[00:39:26] LP: It shouldn’t even be allowed.

[00:39:31] MK: It just was a total violation of the data pixel ratio. It was just putting all of this stuff on there. I do find following some of the defaults, so I mentioned that I turn a lot of them off, things like heavy gridlines, stuff like that, that's not really necessary. Where it tells you like, when you have a table, it puts row one, row two, row three, I'm like, “I don't need to have those numbers there.” I don't need the number of pages because Data Studio is intuitive, people can scroll through something in particular. So, just pulling things like that off.

I find clutter, I guess, is actually probably one of the things that can make it more offensive, is when people don't think through thoughtfully how to clearly articulate information and remove the clutter that's on the page. I think that that is definitely something that is challenging. That would be the case no matter what tool you were using. If somebody's going to create something that is not thinking about data visualization best practices, they would do the same thing in Tableau. They would do the same thing in Excel.

[00:40:46] LP: It’s pretty universal.

[00:40:49] MK: I would say huge behemoth, multi-page reports. But I'm also guilty of it myself. I have things that I have built, that are two and a half years old. And over time, we added more and more to it, and they are massive, but I think you still have to step back and say, “How can I organize this? How can I structure this? Can I put it in sections? Can I title the pages in such a way that it's still intuitive to somebody to know how to think through it?” So, I think those are kind of the places where I see the biggest missteps. And in a way, they're small things, but they're the big things. They are the things that are going to stop people from being able to clearly understand the data, if you're not thinking through it from an end user perspective.

[00:41:38] LP: That's really the key, Michele is, what about them? I think Simon Sinek always said it best, “Make it about them, not about you.” That's why you are absolutely right. The clutter issue is universal. We'll bring that to every tool. I'm still dreaming of a tool that will have a clutter alert, like, “You are reaching your clutter threshold.” We're not there yet. But with technology advances, if Google is listening, you never know.

But we don't always know how to think about it from the end user perspective. And that's why, again, I don't get bogged down by questions of, “Is this too many slides?” Because with every single slide or visual or dot I’m adding, I am asking, “What is this in service to?” Is this in service to comprehension, to elimination, to visual appeal? Which is not a bad thing to go for. People do actually prefer to look at information that's visually appealing, but not at the sacrifice, to your point, not at the sacrifice of the actual information, the accurate, clear interpretation of the information. So, I appreciate that a lot.

[00:42:55] MK: Well, and I have a little bit of an example of that, too, because Data Studio has a limit to the number of elements that you can put on a page, and I have myself hit that limit. Ironically, sometimes it's not because I'm trying to put too much clutter on a page. It's because I'm trying to simplify. So, I am doing things like putting a white box over axes that are unnecessary because I already have data labels there. So, they don't need to have both of them. They've already got a guide to read the data. And so, I'm putting something on top, I'm putting text that is like a cleaner simplification, rather than what Data Studio does by default. I’m hiding things. I'm annotating it in such a way and I end up with a lot of stuff on the page. But ironically, it's trying to make it more minimalist. It just takes a lot of things to do that.

[00:43:46] LP: Right. Because you know what's going on behind the scenes, and it's in service to creating something more simplified. Again, it's not about the number of slides, and it's not about the number of objects. It's how things are designed and arranged to make things clear, right? I appreciate that.

[00:44:02] MK: Yes. My very first boss in analytics once said to me that when you're an analyst, you have to think of yourself as an information architect. That's your job; to think about how to put together information so that it's easy to digest other people and that has always stuck in the back of my head. It's always there whenever I'm doing any kind of work.

[00:44:21] LP: Yes, absolutely. I always say it's about the skill and intention. Just like if you're performing surgery, I use this example all the time probably ad nauseam of the scalpel, where a scalpel is this inherently neutral object but if the surgery goes wrong, generally, people don't blame the scalpel, it's the skills gap and intention of the person using it with other variables. So, these are all just tools. These are all just scalpels. It requires our skill and intention for using that.

[MUSIC]

[00:45:02] LP: All right, so we've entered the next segment of the show which is called The Upgrade. The Upgrade is a tool, a resource, a book, a person, something cool that people can check out that you're loving right now, or was really important to your journey as a data storyteller, and people just love to check out. So, what do you got?

[00:45:23] MK: I have two. One is, if you are a Data Studio user, one of the things that I both love and hate about it is that it is always changing, sometimes significantly. So, things that weren't possible a month ago are now suddenly possible because of new features. I would say that keeping track of what gets released is often the difference between being able to wield that scalpel in an intelligent way, because like, “Oh, this is now something that I can actually do.” Whereas before, I was saying, like, “Actually, that's not possible within Data Studio. You can't do it.” So, I would say that that is a good thing for people to be keeping up to date on.

There's also one blog in particular, it's wissi.fr/blog, W-I-S-S-I. His name is Mehdi, and he has some really creative uses of Data Studio, just like really clever stuff. Oftentimes, when features come out, he will be one of the people that is posting about how to use it. So, I think that that's a good place.

If you're just starting out, though, I do have a Data Studio course on the CXL Institute. So, people are welcome to check that out, if they're trying to get up to speed and learn some of the foundational basics.

[00:46:45] LP: Oh, those are two great resources, and CXL is the best. So, I'm sure it's super high quality.

Well, those are great resources. I'm definitely going to be checking them out. We have actually arrived at our final question. So, I want you to think hard here, and imagine this very plausible scenario. You're entering the cooldown of this amazing group fitness class that you're leading, when suddenly you trip and fall into a vortex that pulls you back to the moment you're about to deliver your first presentation. Do you remember what you were presenting about and what advice would you give to that person?

[00:47:24] MK: So, my first conference presentation, I remember vividly.

[00:47:28] LP: We all do.

[00:47:31] MK: I was talking about the way that we did advertising forecasting back at Kelley Blue Book, because we had to sell ad space up to 18 months in advance. We had to literally forecast down to the ad level, to be able to tell advertisers what they were buying. So, there's very direct revenue consequences. And I was trying to explain how we do all of that. I have at one point, in the past, looked back at my slides, and oh, boy, they are special from that presentation.

[00:48:03] LP: I want to see.

[00:48:05] MK: I would probably tell my past self to read a couple books before putting together the presentation, because that would have been a good place to go, and that an internal presentation is not the same thing as something that you're giving externally at a place like a conference. They're very different. But I think the mark for me of what makes a good presenter, versus not, is sometimes just the enthusiasm for it.

Obviously you need to have a certain level of skill. You can't go up there and be talking BS and not know what you're talking about. So, I consider that to be like, the bar is that you know your topic and what you're talking about. But when you actually care about it and are excited about it, you bring the audience along with you. I think, being able to harness that, when delivering a presentation and being able to say to people, I've done lots of presentations on Data Studio, and being able to say, “This is a really cool thing that you can do with it.” It does engage people in a way that no matter how knowledgeable you are, if you're not excited about it, it doesn't come across in the same way.

[00:49:15] LP: That's right. And even if you're a great speaker, and not knowledgeable or you’re a great speaker and knowledgeable without that passion, that passion is really the thing that engages and unites and draws people in. Because people want cool stuff and they want to know that you find it cool when you're sharing these things. That's why I think public speaking skills are such an amazing value add in terms of something for data practitioners to invest in themselves, even though it feels like not a critical skill to have in their toolbox.

That's great advice. I’d probably give myself all the same ones. But Michele, unfortunately, our time has run out. It flew, which means we must have had a blast. I know I did. So, please tell the listeners where they can keep up with you.

[00:50:05] MK: Yeah, I have a blog on the Analytics Demystified site. You can follow me on Twitter. I will certainly confess I have not been as talkative as I have been in years past. Life has gotten away from me a little bit. You can find me, I'm on all of the usual interwebs. So, blog, Twitter, LinkedIn, all of those places. Feel free, if anybody has questions, definitely reach out to me. I think I'm pretty accessible.

The other place that I spend a lot of time that they may want to check out if they haven't already is #measure chat. So, it's join.measure.chat. It's a community of – it's now well over 15,000 analytics professionals. It's a lot of people. But some of the smartest people are there, and whenever I have questions, like that's the first place that I go, and you can find me there as well.

[00:50:54] LP: That's the Slack channel?

[00:50:56] MK: Yes.

[00:50:57] LP: Oh, it's known as #measure chat now. I know, I gotta get back on there, too. It is an incredibly lively resource for sure. Great, well, all of the links that she mentioned will be available on the show notes page for this episode. Michele, what can I say? It's amazing to see you after the last few years, and I'm really hoping that our paths cross again at a future conference now that things are actually starting to be on real stages again. I appreciate all the value you bring to this field and inspiration as a woman especially, and I hope to get see you again soon.

[00:51:32] MK: Thank you, you too. And thank you so much for having me.

[END OF INTERVIEW]

[00:51:44] LP: Okay. Well, that was an interview 10 years in the making, and it didn't disappoint. Michele's breadth of knowledge is unparalleled in this field, and I'm so glad you got to enjoy it. It's now a real milestone to have hosted both Kiss sisters on the show. The other is Moe Kiss, host of Analytics Power Hour. So, if you haven't checked that out yet, you must.

So, to catch all of the links to the takeaways and resources mentioned in this episode, visit the show notes page at leapica.com/083. If you'd like to connect, don't be shy and reach out to me on LinkedIn or Twitter, and be sure to send a connection invite with a note mentioning the show. I love to meet listeners and I respond to every message.

I'll leave you with today's data presentation inspiration by Jagat Saikia, and that is, “The effectiveness of data visualization can be gauged by its simplicity, relevancy, and its ability to hold the user’s hand during their data discovery journey.” I love this quote, and I think it's so applies to dashboards. I believe a well-designed dashboard puts your decision-makers in the driver's seat of the most important and basic decisions in their business. So, here's to creating an effective data visualization in dashboards to guide them on their road to business victory. That's it for today. Stay well and Namaste.

[END]