When you think about storytelling, what immediately pops into your brain? Does it involve, say, a princess in a tower or maybe a few dwarves with colorful names?
Most people are probably nodding their heads right now. It makes sense — culture has us programmed to associate storytelling with a fairytale that starts with “Once Upon a Time.”
But telling stories isn’t exclusive to urban legends and animated movies with singing animals. It can also be used to illuminate — wait for it — your data.
On this episode of Comp + Coffee, the hosts are joined by Cole Nussbaumer Knaflic, CEO of Storytelling with Data. They chat about how to get your data story started, how to visualize data for every level of your company, and Cole gives some tips on how you can make your data more engaging in a snap.
For a full transcription of the episode, see below.
Kaite: All right. And we’re live. I am channeling my inner Shawn Lavana today, Bill, because Shawn is not here with us. I mean, he’s still here with us, you know.
Bill: That doesn’t sound good.
Kaite: He’s just not here at “Comp & Coffee” with us today. So today, it’s just me and you, Bill.
Bill: And we’re live.
Kaite: And we’re live. And we’ve also got Cole Nussbaumer Knaflic on the line with us. And Cole is joining us today to talk about storytelling with data. She is founder and CEO of Storytelling With Data, the company, and author of “Storytelling with Data: A Data Visualization Guide for Business Professionals.” And she’s here to really chat with us today about this topic that we’ve kind of been talking about, I feel like, for a while now, about how to incorporate data into your conversations about compensation. So obviously, Cole, your background isn’t specific to comp, but you will be able to help add some, I think, really relevant information to this. So welcome, and thank you for joining us.
Cole: Yeah. Thank you, Katie and Bill, for having me. I’m looking forward to our conversation.
Kaite: Cole, one thing that…when you and I were chatting in advance of this episode, one thing that came up that I think is going to be really interesting to our people is that you do actually have a background in HR. Can you tell us a little bit about that?
Cole: Yeah. So I had the opportunity to join Google at a very exciting time, which was right as the people analytics team at Google was being formed. And so, previously, analytics had been done within HR by an analyst, here and there: a couple analysts in comp, a couple analysts in benefits, and so forth. And I joined shortly after they brought everybody together into a cohesive group, and where we started to really look at data holistically from a people perspective at Google. Which was very exciting, because the team was tiny, and we were learning about things for the first time.
And as you can probably imagine, Google has a lot of data, and we [inaudible 00:02:19] data on everything. And so, we had a lot of interesting data to be able to look at and start to tell stories with, and inform how people decisions at Google…so decisions about employees or future employees, were made.
Bill: It must have been pretty cool and pretty stressful to be doing data analytics on people at Google.
Cole: Not stressful, I think. It was interesting. I remember, my…it was some of my first taste of people analytics. I have a mathematical background, had been in banking prior to that, so I’d always been around a lot of data and doing a lot of interesting things with data. And for me, one of the, I guess, things that you would expect, going in, that maybe I hadn’t fully thought through is just how much these sort of analytics you can do on people is similar to analytics you can do elsewhere. So predicting who’s not going to pay your loan back is actually sort of similar to predicting who’s likely to leave the organization and when, and how can you get ahead of some of those things. So it’s really exciting, and the outcomes you come out of your analytics with are tangible, and they impact people. And so, that’s all very exciting.
Kaite: That sounds very interesting, and kind of a good segue into what you’re here to talk to us about…talk with us about today, storytelling with data. And one thing…I just want to set the stage for our listeners. I think a lot of people who aren’t maybe in a communications background hear “storytelling,” and they think, like, Cinderella and Snow White, whatever, that type of fairytale storytelling, “once upon a time” storytelling. And that’s not the same type of storytelling that we’re talking about, right? There’s lots of different types of storytelling, and I think the core that we focus on here is communicating something to your audience. And in this case, using data to help paint that picture.
Cole: Absolutely. And so, it is this interesting thing, because story, I think people have preconceived notions of what that is. And we think it’s this fluffy marketing tool, “Oh, we should tell stories with our data.” And people will sort of talk about that with a little bit of…I don’t know, skepticism, I guess. But you can actually be very strategic in the way that you use story to communicate. Because if you consider…you know, you sort of poke fun at these “once upon a time” stories. And clearly, that’s not the sort of story we’re going to tell with our data. But when you pause and reflect on, you know, we can all remember a story that started once upon a time. And we can recall that plot for ourselves, and the twists in the action, and the ending.
And so, that’s part of the power of story, is that stories are memorable. And they’re memorable in a way that facts on a slide, or data in a spreadsheet simply are not. And so, when we can combine those two things, right: technical analysis that is robust, but be thinking about, how can we weave that into a narrative, that combination is incredibly powerful.
Bill: Right, I can totally see that. And also, how to get people to remember the right part of the story. And so, sort of get rid of the parts that aren’t relevant to the message, the takeaway that you want in your story.
Kaite: Yeah. And then, using data to help not just to tell the story, but I think, also… From the person communicating this, right, whether you’re the speaker at a presentation, or someone communicating in a meeting, whatever, the data… And I would love to hear your thoughts about this, Cole, is, I think the data is kind of your anchor points, right? The data will move you from slide to slide, or section of your presentation to section of your presentation.
Cole: Yeah, it certainly can do. And the other interesting thing about data is that we can make it visible. We can aggregate it and make pictures out of it…you know, graphs, and allow people to not only hear our story, but to see it. And that’s another powerful way of tapping into people’s memory, right, when you both tell them your story, or you say your words and you show them a picture of that story, a graph that shows the data that helps illustrate the takeaway that you want them to focus on. That helps with memorability, as well, and stickiness of your message, because now, not only can people recall what they heard, but also what they saw. And that visual memory is actually a very fast thing for humans.
And so, when you think about not only showing your message in a way that’s going to be understood by the immediate people to whom you’re presenting, but then, also their ability to recall and potentially retell that message to someone else becomes more likely, as well.
Kaite: Yeah. Yeah, that’s super interesting. And you know what? I feel like somebody who’s not as comfortable in a communication standpoint, maybe they don’t feel like they have the best communication chops on the block, that…I think that the, “Will my audience remember this information,” can easily get overlooked. And you’re right. People now have the span of…a memory span of, I think, six seconds, eight seconds at this point, is what they’re saying, with all the information that we’re overloaded with. And having that visualization is just the key to making someone remember it and actually get the point across.
Cole: Well, and this has always been really interesting to me, because if you think of the typical analytical process, you start off with a hypothesis or a question that you set out to answer, then you gather the data, then you clean the data, then you analyze the data: all that behind-the-scenes stuff, it actually takes a ton of energy and time. And oftentimes, we get through that whole, big behind-the-scenes process, and then we just throw it in a graph and maybe outline some findings, and we stop there. And that graph is the only part of the entire process that our audience ever sees.
So my view is, it deserves at least much time and attention as all of those other steps of the analytical process. But it’s the one that can most easily be skipped, yet it’s the one where all of the good work that we’ve done either succeeds or fails. And so, I think the good news is, particularly for folks who come up from a quantitative background, who maybe have a slight hesitancy when we say this word “storytelling” or “communicating,” that investing in skills there, right…because they’ve already got the quant skills.
And if you can take it to that next level to say, not only are you doing fantastic analysis, but now you can communicate it in a way that’s going to make sense to somebody who’s less close to it, that they’re going to remember, that they’re going to be able to retell, and that investing in those skills is what can really set people apart and help ensure that requisite attention is paid to the good analytics that are being done. So I’m a firm believer that there is a tremendous amount of value to be obtained by analytical work that’s already being done that just isn’t being communicated as effectively as it could be.
Bill: Right. And I think some of that would also be, you need to do a lot of data manipulation, analytics, etc., to come to a conclusion. But you don’t need to show everything…
Bill: …to your audience, because there’s a lot of stuff going on there that is not necessary to get to the point. And when you’re exploring, you sort of go bottom-up in digging around and find all the things that you need to know. But when you’re explaining, you start with the answer, or the conclusion.
Kaite: I was just going to say that, start with the end in mind.
Bill: Right. And then, only… You know, you’re not getting graded on all of the detail. You need to have the backup, but just have the key points out there. Because as you said, to make it memorable, you want to have a single point of focus on a page, or graph on a chart, or whatever.
Cole: Well, and it’s not a college class, right? You are not being graded, as you said. You don’t need to show your work, we don’t need the proof. People are expecting you know your stuff and you’re doing the right sort of robust analysis behind the scenes. But yeah. And it’s an understandable desire, right, because that part of the analytical process takes so long, typically. And…when you look at all these different things and all these different directions that actually don’t turn into anything interesting or noteworthy. And there’s an understandable desire to want to take people through that, because it took a lot of time, it took a lot of energy. You want them to feel a bit of that pain.
Bill: You want to get credit for the work you did.
Cole: Right, right. But the audience is going to be much more appreciative when you step back and really think about communicating… And it’s this paradigm shift, right, because it’s easy to communicate for ourselves, or for our project, for our data. It’s a much more nuanced thing to actually flip that around and say, “I’m not communicating, first and foremost, for me. I am communicating, first and foremost, for my audience, and really thinking about not what are my needs, but what are theirs, and how do I make what I need to happen overlap with their needs?” Because if you can figure that out, that is a successful space from which to communicate.
Katie: We talk about that a lot. So we talk from our audience’s compensation professionals, from their perspective of communication and how, really, like… Communicate in the way that your employees, your managers, your executives…communicate with them the way that they want to communicated with. And it’s going to be different from across audiences, it’s going to take that in-depth understanding of who your audience is and where they go to communicate, and hit them across…not hit them. Reach out to them across…
Bill: Sometimes, hit them across the head with it.
Cole: Yeah, meet them where they need to be met.
Katie: Yeah, yeah. Meet them where they are, and don’t expect them to come to you. And I think the same goes true…or it is true when it comes to what you’re communicating. Like you said, be strategic about the points that you need to make, and what data backs…and facts and information back that up.
Cole: Yeah, it’s interesting. So my team and I, we teach a ton of workshops on how to communicate effectively with data. And it’s part data visualization best practices, part storytelling, and how do you weave that into a compelling narrative that’s going to get your audience’s attention and build credibility and motivate them to act. And what we found over time is, we’ve been spending an increasing amount of time and energy and focus in these workshops, which typically are half-a-day or day-long sittings with an organization, focusing on things that actually have very little to do with the data, that are more about, how do you communicate effectively. And stepping back and thinking about, “What is the main message that I want to get across?”
We do a lot of low-tech exercises with sticky notes, where we’re storyboarding and figuring out what that narrative looks like, and how we might rearrange it, and when data needs to play a part, and how do you pull that, then, all together. And it’s interesting, because a little bit of time planning actually saves you a ton of time executing, because it cuts down on iterations. And particularly, when you can get stakeholder and client input or feedback at that low-tech point, it helps speed up and make more efficient a lot of the rest of the process.
Katie: Absolutely, absolutely.
Bill: I think that there’s a great point in there that I just want to suss out a little bit, of, come up with the storyboarding, as you put it. Or the…like, “what are you trying to communicate” should be the first thing you do in your presentation. And I’m guessing…I’ll make up a number, because I’m good at that. Seventy, eighty percent of the people listening would…
Katie: I think you’re right.
Bill: …would start a presentation with the data and the graphs. And then, they put titles on the graphs, and then they’re done. And I think the most effective presentations I have worked on or seen start with, effectively, a table of contents: what are the headlines, what are the messages, and then you sort of fill in the story in a way that…
Cole: Well, and I think…
Bill: …makes it memorable.
Cole: …it can be useful to start with the data, right, to get to know your data enough that you know what directional it’s going to take you in. But then, to step back from that…
Bill: Right, that’s the story.
Cole: …and then, yeah, to plot it out in words. What’s that? You know, what’s your main message going to be? What are the components that you’re going to use to get people there? And that’s when you can really step back and think about your audience, as well, and consider things like, what biases might they be coming in with? The data you’re going to show, is that going to confirm something that they already think to be true? Or are you trying to debunk something that’s a strongly-held belief?
Because thinking about those things and strategizing, well, how do you deal with that, who’s going to be there in the room, are there influencers who can help, are there people you need to sit with ahead of time, so they don’t derail things: that…all of those sort of conversations can really help set you up for success when it comes to the ultimate presentation. And then, after that is when you come back to the data and say, “Okay. Now, given all of this, when does that mean we bring our data in? How do we show it? What’s the message we want to get across? What do we want people to remember?” And then, given what you know about your data, you can design the graphs that are going to help facilitate that.
Kaite: Yeah. We’re sitting here nodding our heads to everything that you’re saying, which obviously doesn’t translate well over a podcast.
Bill: I’m nodding.
Kaite: And I think I feel like there’s one kind of sidebar that you just touched on, which was understanding who’s going to be in the room when you’re presenting this, too. And I don’t feel like…if that’s something we talk about as much, “we” meaning our team on this side of things, when we’re talking about communication. But you’re right, that’s so important. And I love that you said sit with the people who could possibly derail things in advance. Or likewise, like, your promoters, sit with them and be like, “Okay, I’m going to need you to back me up on these points,” or with the…I’ll call them the detractors, people who could derail things, explaining… You know, that would give you an opportunity to maybe go more in-depth with them or ease some of their concerns that they might otherwise voice and kind of take your presentation off the tracks.
Cole: Yeah. Something else to anticipate ahead of time, because…especially if you’re presenting to a mixed group or you have different stakeholders who need to be in the same room. When you can identify that their needs are sufficiently different, that’s where you want to think about…and particularly if the stakes are high, how do you align things to really line yourself up for success? And one place where those are often different, to give a concrete example, is, somebody in the room wants a ton of detail. They want to go into the data, they want the nitty-gritty. And another person wants the big picture. They want the story, they trust you’ve done the analytics, they don’t want to go into the nitty-gritty. And those are impossible needs to meet simultaneously.
So if you can identify the people beforehand who are going to want the details, that’s where you take them aside and sit with them, right, and go through that, so that they are satiated, or that need has been fulfilled by the time you get to your meeting, so that it doesn’t throw things off track.
Kaite: And I think what you’re saying now, and what we just discussed kind of dovetails nicely into one of the questions that I wanted to touch on with you, which was how to tell a story to all different levels. For us, it’s…within an organization, it would be our three different…primarily, three different audiences that most of our listeners interact with, which is employees, managers, and executives. And I think that some of the points that you just hit on would help them do a good job of communicating with those various audiences. You may not always have those three levels of people in the same room for the same presentation.
What about when…? What about…? Let’s say that they’re doing a presentation to execs individually, managers individually, and employees individually? How should they use the data to tweak a story, in that case?
Cole: Yeah. That’s a great scenario for thinking through strategically, how do you deal with this so that it can be successful in these three different scenarios? Because I think what happens most commonly…or what I see what happens most commonly is, we design one presentation deck, or one communication, and it’s trying to meet all of those needs. And what happens is, when we try to meet so many needs simultaneously, we actually don’t meet any one of those sets of needs as well as we could. We actually just allowed ourselves to have separate communications.
So one thing that I always recommend is, if you find yourself communicating to a really wide audience or a mixed audience, question the assumption that you have to communicate to everybody simultaneously. Because if their needs are sufficiently different, there can be value to breaking that up and doing things differently. I’ll call on a Google example here, which is, when I worked in people analytics at Google, our main audiences were our sales organization on the one hand, and engineers on the other hand, right? These were the people that we needed to convince of our findings or what actions we thought needed to happen. And the needs of those groups are totally different, when you step back, I’m realizing here.
But salespeople, their general view was, “Leave us alone. We’re the ones out there making the money. Don’t bother us.” And the engineers, on the other hand, had this endless desire for detail. They wanted to be convinced that our statistical methodologies were sound, that we’d done the right sort of methodology for our forecasting, that we checked all these corporate cases, and so forth. And of course, when you say that out loud, you can’t meet those needs simultaneously. But it was actually after trying to, a few times and failing that we thought, “Okay, we’ve got to revisit how we do things here.”
So in that case, we said, “You know what? We don’t actually have to communicate to everybody all at once. The salespeople, let’s leave them alone.” We won’t touch them until we have something concrete we need them to take action on, and hopefully, a compelling reason for them to do so. The engineers, on the other hand, we’d have to get them involved, sometimes as early as we were designing the study, so that by the time we got to the output, they were bought in, and they knew our methodologies were sound and influential. And engineers could help influence those around them.
So if you find yourself facing a mixed audience, always question whether you have to communicate to everybody simultaneously. Now, I say that, recognizing that sometimes, we do. We can’t make different presentations for different people, for whatever reason. And so, when that’s the case, then you want to consider how you’re going to be presenting. Are you there live? Or you can use verbal cues to get the right people to tune in at the right time. Or is it something you’re sending around? And then, you want to think about how you structure your documents, such that people can easily find and turn to the sections that are going to be most relevant to them. So again, it’s coming back to thinking about your audiences, right? Where are you going to be meeting them, and how do you need to do that, to try to make it successful each time?
Kaite: Yeah. And I think that starting with the audiences, and always questioning when you’re going through a communication, whether it’s an email or a presentation or whatever it is, “Is this serving my reader?” I’m a writer and a marketer, me personally. And that’s kind of how I structure everything, is, “Is this the best sentence? Does this seem like…? Does this start with the reader in mind?”
Cole: Yeah. And I think optimizing, as well. Because we… So in our workshops in the book, which I can talk more, as well, we cover specific lessons for how to do this well. And it doesn’t mean you do every single one of these, every time you touch data. You’d never get anything done. But it means being smart about when and how you use your time when it comes to this stuff. When does it have to be perfect, because it’s going to the executive team and it’s a high-stakes thing? Versus when can it be quick and dirty, and that’s okay, and it’s going to get the job done. And being really explicit about which of those scenarios, or where on the range in between them you are, for a given thing.
Katie: Yeah. Yeah, I think that’s all really important. So how…? This might be a generalized question here. But how could somebody who may not, you know, at service level, consider themselves to be a communicator? Maybe that’s not something that they would consider to be in their wheelhouse, or something they’re super comfortable with. How could they get started, easily wrapping a stronger story around the data that they already have, the data they’re already compiling?
Cole: Yeah, great question. And I think the first step is to not just put data out there, which, it’s an easy thing to do. We do our analysis, we make the graph, and we maybe outline some findings, and we stop there. And the challenge in stopping there is, to your point, your audiences are faced with a ton of data every day. So when we give them more data, it’s easy for the conversation to be, “Oh, well, that’s interesting.” And then, they move on to the next thing, because you’ve missed that, what, six seconds or whatever. You talked about it earlier. [Inaudible 00:23:14]…
Bill: “That’s interesting,” is generally a sign of something that’s not interesting.
Kaite: Yes, yes.
Cole: Right. Or they ask you for more data. That happens a lot, too. And then, you get into this death-by-data cycle. And I think the way to avoid both of those scenarios is to take it a step further. Not only, “Here’s the data,” but, “Here, audience, is something you could do with this data. Here’s a decision you could make, an action it might inform, a conversation that you could have, some options to consider.” Because that gives our audience something concrete to react to. It invites a response. And even if the audience disagrees with what you’re recommending or putting out there, it starts a conversation. And it’s a conversation that often gets missed when we stop at simply showing data.
So I think, for me, the first step to somebody who is in an analytical role but doesn’t consider themselves to be a communicator is, don’t just inform. If you’re thinking of your role as to inform, that’s actually a much…that’s too passive, and you’re leaving too much value on the table. Because if you’re the one analyzing the data, you likely know it better than anyone else, which puts you in a unique position to help others derive value from that data. And the way we do that is by helping them understand something so that they can do something differently, or so it can inform something that has been done a certain way and should stay the same, so that we are not just looking at data to look at data. We’re looking at data to influence smarter decisions and good actions.
And so, the first step is to be clear on what you think those should be, or give your audience a starting point to react to, so you can get additional context and be recommending smarter things, as you learn and move forward.
Kaite: So being… You know, we talk a lot, Bill, about being proactive, not reactive. And I think that everything you just said backs that up. We’ve done a lot of…we’ve talked a lot about being a comp strategist, being more strategic and more organized in your role. And I think everything that you’ve just said, Cole, is exactly in line with that.
Cole: Tactically, right? Because it’s like, make a recommendation. But tactically, I think that the way to do that in your tools and with your graphs, two areas I’ll often recommend is, be really smart in your use of color and your use of words. Color, used sparingly, is one of your most strategic tools to draw your audience’s attention to where you want them to look. So you can think about, if you’re designing a slide or you’re designing a graph, all the details are there for a reason. But there are some data points, or some parts of that, that are going to be more important than others.
And you know…as the person who put together that graph, you always know what it shows and where your audience is meant to look. The challenge is, those other people don’t live in your head, right? They don’t have that tacit knowledge. And so, we actually have to take explicit steps to direct our audience’s attention to where we want them to look. And one easy way to do that is to use color really sparingly, because color, when contrast is sufficient, is a huge attention grabber. So you can think about something you might normally do, and it’s kind of colorful. Try making it shades of gray, and then use black or a single color or two where you want your audience to look first. And what this starts to do is create visual hierarchy.
Coming back… I forget, Bill, if it was you or Katie who mentioned this, six seconds of attention, right? Don’t know if it’s six seconds, but the point is, it’s short. You don’t have people for long. So the more you can do to make the information you’re putting in front of them scannable, where they can spend a few seconds, their eyes glossing over, and their eyes, when they do land on the most important things…because that’s how… If it’s something that piques their interest, they’ll pause more, and then look into more of the detail. So using color sparingly to direct attention in that way. And then, using words wisely to make why you’re putting their attention there make sense.
So words that describe the takeaway, or…and minimum words that make the graph intelligible, right: titles and access titles and such. But taking it a step further than that and putting into words what you want someone to see in the graph. So color and words, used well, can overcome a lot of other issues and just make the analyst also really thoughtful about, “Well, where do I want my audience to look? What do I want them to take away from that graph?” So if you can’t articulate that, you shouldn’t be showing it in the first place.
Kaite: Good point.
Bill: I love the color point, because I think most people…sort of like we were saying, you get your data, and then you just plop it down onto a slide, because you’ve done all this work, and you just want to move on. Also, when you do it in Excel or pretty much any other software, your default is the Crayola six-pack. And it’s so much color, you don’t know what to look at, you don’t know what’s [inaudible 00:27:57].
Cole: Right, because everything’s competing for attention.
Bill: Right. And so, you want one thing to win the page, one concept. And you want it highlighted with color, and then also, like Cole said, highlighted with words.
Kaite: And the text, too, different text sizes. In the same way that you’re using color, same principles apply to how you’re treating the text itself on a page. And also, Bill, you just said it. And I think this is…so many people are guilty of designing presentations specifically, where it’s, “How much information can I fit on one slide,” not… Like, let…give it room to breathe. Let it…one big point per slide, like you said, Bill, not three, four points. It’s okay to have 50 slides versus 25, if there’s one solid point for people to kind of mull over per slide.
Cole: Yeah. And we’ll use animation often. And it’s a strategy that we use to illustrate, particularly if you’re building up to something that going to be complicated or dense where, instead of starting with the complicated, dense thing, you actually build it up piece by piece. So it can be…you know, if it’s a graph, it’s can be useful to just start with just the skeletons: the bones of the graph, the axes and their labels and titles. And then, you can set up for your audience what you’re going to be looking at. You can talk through what the data’s going to be before you show them the data, which both helps ensure that they’re paying attention, because once you put the data up there, people focus on that and may not hear those words any longer.
But then, also can create some anticipation on the part of the audience of, “You know, you’ve got this big, empty screen. What’s coming next?” And then, figure out, when does it make sense to layer on one data point at a point, or one data series at a time. Or if you’re on a slide, one set of images or words at a time. And in doing so, you can actually build up to something that’s quite complicated, but not have it feel overwhelming, because you’ve taken people through that buildup, step-wise. And that can sometimes be a useful strategy.
Bill: Yeah, I think… I’m laughing because I think that’s great. And one of my former, former bosses…shout-out to Heidi Willis-Turner who’s watching, used to joke when she was meeting with CEOs that they would flip through a binder, and they would flip ahead. And she would be like, “If I were a schoolteacher, I’d be whacking your knuckles with my ruler. Don’t go ahead.” But it shows that they want to know what’s next. And then, you’ve, at least at one level, won, because you have them caught up in your storytelling.
Cole: Well, and that’s where I think…coming to one of these points we raised earlier of, start with the “so what,” start with the message. Because then, you’ve given them a bit of a blueprint of where you’re going. And the conversation that happens along the way can be more robust because of that.
Bill: Exactly. Yeah, I hate that mystery, what’s the last page going to be, in this context.
Cole: Right. [Inaudible 00:30:47] run out of time before you get there, or something else happens that throws it off.
Bill: Right, or that discussion goes way off on tangents that are irrelevant. But people don’t know they’re irrelevant, because they don’t…
Cole: They haven’t gotten there yet.
Bill: …know where you’re trying to get to. Yeah.
Kaite: I also think that what you’re saying, revealing it piece by piece, also helps with making it more memorable. Going back to, all of this needs to be memorable if you want your point to really come across.
Bill: How do you feel about pie charts and multiple pie charts?
Katie: Oh, boy.
Cole: We actually…we posted… Mike on the team posted a blog post on our site the other day that had 45 pie charts that we were actually talking about as a good example. So I’ve softened my views on pies over the years. I’ve done talks called…titled things like, “Death to pie charts,” and such. I think it’s fun to be a little provocative, sometimes. But I’ll back up and say no chart type is inherently good or evil, that pretty much every graph has its perfect use case. The problem is, you veer too far away from that perfect use case, and things can get pretty confusing pretty quickly. And so, pies, what they’re meant to show is a piece of a total.
And so, where they do that well is where one segment of the pie is really big, or where another segment of the pie is really small. Where they break down is when that’s not the case, when you have 100 different segments, or when multiple different segments are similar in size. Because when we are looking at pies… And people are drawn to circles, by the way. It’s this weird phenomenon. We’re drawn to them, but our eyes actually do a really poor job of ascribing quantitative value to them, meaning they’re actually not great representations of data. And pies, in particular, we are encoding the quantitative values, where we’re trying to read them based on area. And our eyes aren’t great at doing that.
And so, in the case where you actually want to allow people to compare segments of the pie to each other, that becomes a difficult thing, and particularly difficult to do with any level of specificity, because we’re asking our eyes to do things that are hard. Whereas if you can identify what it is you want your audience to compare, you want to think about, how do you put those things as physically close to each other and align them to a consistent baseline, to make that comparison easy? Because our eyes are very good at assessing relative length when the length of the things we’re assessing is aligned to a common baseline.
And that sounds really complicated, but what I’ve just described is a bar graph. Our eyes are good… If you picture a bar graph, it can be a vertical bar graph or a horizontal bar graph, particularly if it’s sorted. But even if that’s not the case, you can use that visual to very easily and quickly pick out which category’s the biggest, or which is the smallest. And also, because of this alignment to a consistent baseline, you can easily see the difference between categories. And so, bars become really useful things for a lot of assorted data that we typically want to show on a day-to-day basis in a business setting.
Bill: I agree.
Kaite: You’re on the same page about pies?
Katie: I actually have a question that just popped into my head for you, Cole. Let’s say that I have to communicate something, I have to communicate some data to one of my audience segments that I communicate to. Let’s say it’s Bill, let’s say it’s an executive. And I don’t have the time to go really in-depth, it’s got to get out to him really quickly. I don’t have time to go in-depth in terms of the design or anything like that. What’s one thing that I can do that makes sure that my story and my data get across in a way that’s as quick as possible?
Cole: Yeah, great question. And so, landing on the right graph is something that takes time. When you’re looking at data, and particularly…well, I guess both to understand it, as well as to communicate it to someone else, it can be useful to look at different views of the data. Because any data can be graphed a ton of different ways. And it’s by looking at these different views that we get to know our data better and get to see different nuances, and where we can consider which one of these views is going to work best for our audience. And so, that can take time.
In the instance where you are constrained with time, and maybe don’t have the time to do that [inaudible 00:35:06], or you already know what the message is, you don’t feel like you need to look at your data more, that’s where I come back to these pointers on color and words. Color, used sparingly to direct your audience’s attention to where you want them to look, and words put around that data that tell your audience why you want them to look there: those two things alone will go a tremendous amount of way where, you know, maybe it’s not the perfect graph type. But if you can direct attention in that pie to where you want people to look and it gets your point across, it might work well enough.
And it’s coming back to these ideas that we talked about as well, of considering our audience and optimizing, given who they are and how important it is. Are the stakes high, or is this just a question that somebody posed that they need the answer to, and you’re not sure what they’re using it for? And being thoughtful about how you approach each of these scenarios, so that you’re optimizing your overall time, to be able to go deep and spend more time when it’s going to be warranted, but not do that on every bit of data you touch.
Bill: And I would think that you would also want to make sure that in the time limited and also scope limited, meaning one or two people you’re presenting it to or giving it to, pick your words carefully. Because that’s something you can add to the image, to the graph, that help people make sure they understand. They’ll read the intro sentence and the title with the picture, and they’ll understand the whole thing is one concept.
Cole: Yeah. And I think anytime you’re communicating with data…and we touched on this a bit earlier. But be clear about what it is you want your audience to see, what you want them to take away. And don’t assume that someone else looking at the same graph is going to get there on their own. That’s where you want to put it into words, because there’s also been some interesting studies done about the priming effect of words, which sounds obvious when we say it. But there have been studies that show when you title a graph with the takeaway, people are more likely to remember the takeaway.
Bill: Totally makes sense.
Kaite: I mean, it makes complete sense, though. I’ve never heard that. I love… That’s a really good takeaway, easy to apply, for our listeners, to what you’re working on.
Bill: And I think one other thing that we saw on one of your “Cole’s Blogs” or somewhere on your website that I…sort of like the limit, your use of color, is also… I thought this was very clever, I haven’t seen a whole lot of this, is, limit what data points get labeled. You had some bar graphs where you only had the number, the length of the bar, labeled on 2 or 3 bars, even though there were, let’s say, 10 on the page. Because the two or three were the ones you were calling attention to.
Kaite: And that’s, like, the benchmark for the others, right, you know, if one’s 10, and the others are less, then they’re all…or lower than that, then they’re all less than 10.
Bill: Right. And you can see it, but it’s like, “Oh, here. I see exactly what she’s calling attention to, what I’m supposed to be…”
Cole: Yeah. Well, and it’s another visual cue, right?
Bill: Right, exactly.
Cole: So implicitly, what that says is, “Hey, audience. This data is so important, I even put the numbers on there to help you remember that.” Because we have all of these attributes, pre-attentive attributes they’re called, at our disposal for helping direct attention when they’re used strategically and sparingly. So these are things like position, right, where you put something on a page. Do people encounter it first, or do they encounter it last? Color is a big one, as we’ve talked about. Motion is actually a super attention-grabbing pre-attentive attribute. So you have to be careful with motion, because that’s also easily the most annoying, right?
Cole: But yeah, size and color and position and adding labels, right? These are all things that you can do to help your audience understand what’s more important, that they should pay more attention to. And if you can do that, right, if you can get your audience looking where you want them to look, that’s part of the…you’ve already partly won, right? And now, you can get to them why they’re looking there and what they’re meant to do next, that is successful communication with data.
Kaite: So Cole, you obviously are an expert on this. I think that this has been an extremely informative conversation. If our listeners want to learn more, where can they find you? I know you’ve got a book coming out in two months, right, in [inaudible 00:39:17]?
Cole: Yes. So “Storytelling with Data: A Data Visualization Guide for Business Professionals” is already out there, and is a great primer. It’s an overview of the foundational lessons for communicating effectively with data in a business setting. The next book is called “Storytelling with Data: Let’s Practice.” And it follows the same topics as the original book, but with a real focus on hands-on practice. And so, each chapter is designed as three different sets of exercises. The first set of exercises within each is called “Practice with Cole,” where I pose a canned example. It’s a scenario or a question or some data to consider that you’re meant to work through on your own.
But then, I also work through my solution as a way to show just a ton of corner cases and more examples and insight into the design thought process behind the scenes when creating graphs and communicating with data. Then, the second exercise section within each chapter is “Practice on your own,” where it’s the same sort of examples, but without any prescribed solutions. And this is for the individual who just wants more practice on their own, or managers who may be wanting to assign things to their teams to help develop skills, or for the university instructor who’s teaching from the book. We have over 100 universities around the world that are teaching from the first book, so this would be a good resource for them, as well.
And then, the final exercise section within each chapter is “Practice at Work,” where it’s, okay, you’ve done this in theory, you’ve done it with some canned examples. Now, let’s take a project that you are facing in your day-to-day and break it into component pieces, and step you through things to be thinking about and questions to be asking, and when to get feedback and who to get feedback from, and really bridging any gap between the theory of, “Here’s how we could do this,” and the practicality of how you can do it with your data, with your projects, with your audience.
And so, I’m very excited for it, and for having, I think, something next for people who go to a workshop, or they read the first book and they get it. They get that doing this well is a skill that has to be invested in, but maybe are finding some gaps between that and being able to practically apply some of the lessons that are covered. So again, that’s called “Storytelling with Data: Let’s Practice.” It is available for pre-order across the normal book venues, Amazon and such, now. And it’ll be officially published in October of this year.
Kaite: Fantastic. We’ll put…I’ll mention it in our notes, as well…
Kaite: …for this episode. And also, people can find you at storytellingwithdata.com. Bill mentioned…
Kaite: …that he was already checking out your blog. We found that really helpful.
Cole: Yeah, there’s a ton of info on the website: resources, blog posts, where we do a lot of makeovers before and after. And so, we always try to always post the underlying files, so if anyone wants to poke around and see how stuff was done. So you can certainly check that out for resources as well. We have a LinkedIn page where we post daily tips. And also, we’re active on Twitter and Instagram, and that’s @storywithdata.
Kaite: Fantastic. And Pay Factors has also done a lot of work around communicating compensations, specifically. So if you’re listening and you want to check out some of our resources on communicating to all various levels in your organization about the value and mechanics of pay, shoot us an email, email@example.com, and I’ll get [inaudible 00:42:39] resources over to you. Cole, thank you so much for joining us on today’s episode. It was really a pleasure to have you on, and I think that… Like I said this whole time, I think this is going to be a really interesting and useful topic for our audience. I hope it is.
Bill: And an ongoing one.
Kaite: Yes. Oh, yeah. Definitely ongoing. The conversation on this, I think our listeners have seen it woven into most everything that we’ve talked about.
Bill: Everybody’s getting more and more data, and need to do less and less with it, if you know what I mean. [Inaudible 00:43:08].
Cole: They need to show less and less of it, right?
Bill: Exactly. Thank you.
Kaite: Yeah, yeah, yeah.
Cole: Katie and Bill, it’s been a super good conversation. And my final words for listeners would be, the next time you find yourself needing to show data, take it to that next step, as we talked about, of not only showing your data, but thinking how you can weave that into a story and use that to drive understanding and action.
Kaite: I love it. I love it.
Bill: The end.
Kaite: Yeah, the end. Thank you, Cole.
Bill: Thanks, Cole.
Kaite: Thanks, Bill.
Cole: Thanks you both.
Kaite: All right. Over and out.