Brent Dykes sits down with host Jesse to discuss and explore the skill of data storytelling. Brent is the author of ‘Effective Data Storytelling - How to Drive Change with Data, Visual, and Visuals’.
He believes - and has witnessed - that being able to effectively communicate with data is a necessity in organizations of all types and sizes. Brent helps data teams around the world master the art and the science of data storytelling with frameworks and techniques to help craft compelling stories with data.
Welcome to the Soda Podcast. We're talking about the Data Dream Team with Jesse Anderson. There's a new approach needed to align how the organization, the team, the people are structured and organized around data. New roles, shifted accountability, breaking silos, and forging new channels of collaboration. The lineup of guests is fantastic. We're excited for everyone to listen, learn, and like. Without further ado, here's your host, Jesse Anderson.
Hello and welcome to the Data Dream Team podcast. My name is Jesse Anderson. My guest today is Brent Dykes. He is the Founder and Chief Data Storyteller at Analytics Hero. Before he founded his own company, he was at Omniture, Adobe, and Domo. Welcome to the show, Brent. Would you mind telling us more about yourself?
Hey, Jesse. Great to be here. Thanks for the invitation to join your podcast. Yeah, I have a marketing background, so I started in marketing and then got drawn into analytics and then spent eight years in consulting, in analytics consulting with Omniture and Adobe. And then I did four years as an evangelist for the analytics, the Adobe analytics platform. And then I went over to the BI world and got out of just marketing analytics and focused on other functional areas and other industries that I didn't have the chance to work with at Adobe. And so that was a really good experience, just broadening away from just the marketing analytics. But yeah, I spent a lot of time working with Fortune 500 companies around the world and evangelizing analytics and data.
It's kind of funny hearing your background on analytics and it's starting all with marketing where you kind of think maybe that people are thinking here on this podcast. Oh, he is going to talk to us about marketing. No, no. Marketing done the right way, in my opinion, probably your opinion is it's all numbers based. Is that your take as well?
Yeah, I mean, when I first started in marketing, probably wasn't as numbers centric. In fact, I had a friend of mine who also graduated in marketing like I did, and she went to a different university and she was talking to her professors and he said, the one way to make the least amount of money in marketing is to go into marketing research, which was the only numbers based kind of part of marketing. But that has since changed. I mean, today, I would say marketers have a wealth of information, maybe overloaded with too much information and data on their campaigns and different apps that they're running and different programs that they have. So there's lots and lots of data there.
Good. So Brent, we've had all kinds of people on this podcast: the evil mad scientist, big data engineers, founders, all kinds of things. Even the godfather of data literacy who's a friend of yours, but we've never had a chief data storytelling officer. Tell us, what does your day look like? What's the interview process and how does somebody get that job?
I'll start with their last question. The way you get that title is by starting your own company and you can pick whatever title you want. So just because I have a consulting business based around data storytelling, I figure I might as well make myself Chief Data Storyteller. But in terms of a typical day, I'm working with clients preparing workshops for them. I have some contracts where I'm coaching individuals on how to do data storytelling or teams, individuals on teams. Also I spend a lot of time working on content. So I'm very active on LinkedIn, writing blogs, currently working on a white paper that I'll be releasing soon. And then the rest of the admin stuff that a small business person has. So I'm accounting, I'm finance, I'm marketing, I'm web development. I'm all of the things that go into running a small business so many hats.
I feel your pain with the many hats. It's interesting that when you start your own business, you think, oh wow, I can spend a hundred percent of my time on this data storytelling that I love. And it turns out there's a lot of extra stuff that has to be done that isn't what you thought it would be.
Well, good. So you introduced yourself as the chief data storyteller, let's go deeper into this. What is data storytelling? And to do that, you wrote a book called Effective Data Storytelling: How To Drive Change With Data, Narrative and Visuals. This was published on Wiley in 2019. Tell us, what is data storytelling?
Yeah. I view it as a structured approach for communicating insights to a targeted audience using narrative elements and explanatory visuals. So in my book, I talk about three key elements, the data, the narrative, and visuals, and how those elements combined to help us tell stories or communicate our insights that we have more effectively to a targeted audience, as I said.
So this narrative, what do you mean by a narrative?
Narrative, yeah, I think that's one thing that gets kind of overlooked a lot. I think there's been a lot of emphasis on the data visualization aspects of data storytelling, but the narrative is really, it's a combination of things. It's how we structure our story. So if you think of a plot, right. So there's a climax, there might be an inciting incident that launches the story. You're going to have rising action. You're also at the beginning of that, you're going to have an exposition or kind of setting that kind of sets the stage. And then after you have the climax and there's a resolution and kind of conclusion of the story. And so the same thing applies to a data story in the sense that we have to organize our content in a certain flow. So you're going to have what I call the hook, which would be the introduction to the story.
You're going to have the climax, which I call the aha moment. Right? So we have an insight that's driving the data story. And then a lot of times when we're telling data stories, not just going to give people the insight and expect them to act on it, we're actually going to flush out. Okay, we're going to look at how you act on this insight? What does it mean to the business? What does it mean to you? What options do you have? What do we recommend in terms of next steps? So a lot of the narrative is just how we connect all of the findings, the key points and the data, the data story together into this structure, this flow that basically leads or guides the audience through the numbers in a meaningful way that just, it's not really to extend it out. We want to be concise.
We want to be to the point, but we want to hit on the key data points. They're going to help them make a decision. So that's a key part of it. And then if we break it down to the individual data scenes where we have, maybe if you think of it like a slide. You may have multiple scenes in a data story. And usually, those data scenes are comprised of data visualizations, right? So we'll have a data visualization in there. But then, in addition, we'll have the title of these scenes and that's where we start to convey the message of what the audience should take away from the data visualization. We may have annotations and other things. So it's a mixture in terms of a narrative that is going to be that structure, but it's also going to be the actual telling of what's happening in those scenes and explaining it to the audience.
What would you suggest someone do if they can't find a narrative in their data?
I would say the thing that you need to do is kind of start looking for an insight. I think that's the key thing. I don't think we're looking necessarily for narratives. I think we're looking for insights. And then once we have an insight, that's when we then, okay, we step back and say, okay, what is the narrative that we can build around this? Now often when we're doing an analysis of something, there's going to be an existing narrative in place. We may not realize what it is, but people are going to have a perception of how the business works. Like say, we're analyzing some aspect of the business and people are going to say, oh, this process works this way and this is why it works that way. And they have an existing kind of narrative that they built up around that process.
Well, maybe our analysis reveals that the process is broken and there's something wrong with it that needs to be fixed. And so, now we're going to have an insight that challenges the existing narrative. Now there may be other data stories where maybe we have an insight that actually supports the existing narrative or strengthens the existing narrative. So there's different ways that we can approach it, but there's always going to be a narrative. And when we have our insight, we can decide whether it supports the existing narrative or whether it challenges the existing narrative.
Sometimes when I'm talking to both teams, perhaps teams that aren't used to doing analytics, or perhaps they're CEO or something like that, they have some sort of overload from what's given to them. They get this cognitive overload. How would you prevent cognitive overload in this data storytelling example?
Yeah, I like to look at the cognitive load theory that was introduced in the '80s. And basically there's three forms of cognitive load there's intrinsic load, which is basically the difficulty or the complexity of the topic that you're talking about. Some things may be very simple. Some things may be very complex. And so that's the intrinsic load. There's extraneous load and that's really things that we use to kind of present the information to an audience, which may not be directly related to what we're talking about. And so there we classify them as extraneous and then we have Germane load and that's actually the stuff that's really good because people, when we have Germane load, that means we're actually working on understanding and processing the information that's being passed to us. So with the intrinsic load, we can't, if we're presenting on something that's complex, we can only simplify it to certain levels.
So it's really about managing the difficulty. And the example that I use in my book is I say, I could be teaching you how to fold a paper airplane, or I could be teaching you how to fly a jetliner. So one of those two is going to be much more complex than the other. And so with the jetliner example, if we're teaching somebody to fly an airplane, then we're going to need to manage how we share that information. It's going to be naturally complex, right? Intrinsically it's complex, but then we need to chunk it. We need to break it down into ways in which it can be more digestible for the audience. Now with the extraneous load that is just content that is not really needed. And so a lot of times what we do with our visualizations and different things, we want to simplify them, remove items, information that's not essential and minimize anything that really doesn't contribute to our story.
It doesn't help us to deliver the message that we're trying to convey to the audience. So that's a key step. We can remove a lot of content that just really isn't needed. If we step back and evaluate whether it's really contributing to the story or not. And then the last thing is maximizing the germane load. So we want people to join us and really contribute or participate with us in this data story. So that may be where we invite the audience to really think through how we've interpreted the numbers. There can be certain strategies that we can use to kind of engage them a lot more deeply and really think through the meaning behind the numbers and maybe even humanize those numbers and really connect with the audience. So there's different ways that we can manage the content of overload. But the main thing I would say is one of the easiest things is just removing a lot of that extraneous information that really gets in the way of a message. And doesn't contribute to understanding on the behalf of the audience.
That's excellent. Excellent advice there.
I know in the past, I've seen you talk about how we humanize our data. How do you go about humanizing your data?
Yeah. I think it's a simple process of looking for the human side of the numbers. I think a lot of times when we look at our data, we maybe lose sight of the people behind the numbers. And so I think the key thing about humanizing the numbers is bringing those people out and putting the data in a context in which we are associating with the people behind the numbers. So an example of this would be rather than talking about cars, if we were analyzing automotive data, we'd talk about the drivers, the people behind those cars and the vehicles. Maybe we're tracking a lot of IoT data, but at the end of the day, it's about the people driving those cars. And so we reposition the numbers around drivers as opposed to just vehicles and technical data that we're collecting from the vehicles themselves.
So you're talking about much more than, let's say personas. So in marketing, you'd say, there's this persona, they're this and this and this. And they have all these characteristics you're saying to do, go even further beyond that.
Well, I think we can take a data-driven approach with building our personas. And so that's something that I think can be really valuable because we can look at the numbers and then we can say, Hey, this is this kind of individual. These are the people, these are what they look at, or these are where they're coming from. These are the characteristics. Maybe we have qualitative data from them in terms of, or demographic data. Any way we can bring it to life and just make the customers or the employees or the voters or whatever data set they're looking at. Let's bring those people to life. I think we can do the same thing with our data stories.
How do you recommend teams be organized to get the most value from data?
In terms of organization, I mean, I've worked with a lot of different organizations. I've seen the centralized approach work well for smaller organizations. I've seen decentralized not work so well. I've kind of settled on the hub and spoke being the most effective where you have a smaller centralized team managing things. And then you have the spokes where the individual analysts or data team is kind of embedded in those different functional areas or those other organizations. And the key thing is that, as I work with different organizations, if you have that centralized model and you're a little bit disconnected from what's really happening on the front lines and these functional teams, I've seen some of those functional teams get frustrated that the centralized team doesn't seem to understand what their needs are or understand their areas of the business that well. And so then what do they do? They start hiring their own analysts. And then that gets kind of messy. And so I think the hub and spoke for me would be the model that I think is the best that I've seen working at large organizations.
Does anyone on that team have the title ‘Data Storyteller’ or is it a specific title that usually has that?
Yeah, that's a great question. One that I get quite a bit right now, because really we're in the beginning stages of data storytelling. It's not like you have people who've been doing this for... Well, I mean people would argue, I have been doing this. I have been communicating the results. But in terms of everything that I'm talking about in terms of beyond just reporting, I don't think we've actually been doing as much as we think we have. Now in the initial start of rolling out data storytelling, I would actually recommend maybe getting one or two or a few dedicated people or identifying within your team who does this well, or has a natural affinity to do this well and who can be kind of designated as dedicated data storytellers. And they're going to really lead the way and start to shape this culture of data storytelling within your organization or within your team. But eventually I see this being a skill that everybody needs to do.
And eventually, you know, if you have some of those initial people who have a dedicated role, what their goal is to be personal trainers, basically coaches, mentors leading the way in terms of what it means to tell data storytelling, and almost like evangelists, right? They're evangelizing this skill. And then I think it becomes a very common skill because if we are democratizing the data to various teams and roles and functions, why wouldn't they have insights that are... If they're empowered with the right tools and training to find insights, and when they need to share them with others, they're going to need those skills of data storytelling. So I think eventually over time, business teams will need these skills, not, not just the analytics team, but definitely on the analytics team, I think will be a skill that analytics professionals, data science professionals will need.
What is the thing you're the most proud of about your book?
Couple of angles on that question. I would say from a content perspective, one of the things I really enjoyed when I was working on that book was looking into the Hans Rosling. He produced, before he passed away, a really famous segment with the BBC called the ‘Joy of Statistics’. And within that, he had a short four minute segment where he looked at 200 countries over 200 years and they visualized the information, this kind of minority report kind of view. And I was able to talk to the production company behind that and really get a behind the scenes kind of view of how that came together. And that was really special for me because I kind of view Hans Rosling as one of my heroes, in terms of somebody who was able to do good data storytelling.
Now, in terms of what am I most proud of? I think when people come to me personally and say, "Hey, your book helped me. Thank you for writing it." That's the most rewarding by far, whenever you have a reader who comes to you and thanks you for writing your book.
Yes, everybody, when you enjoy their book, tell the author about this. Do it on LinkedIn, do it on Twitter. We enjoy hearing our work. We didn't just write that book to send it out into the ether. We actually do like hearing that. So please reach out. So let's say, I'm a Chief Analytics Officer and I'm thinking, oh wow. This data storytelling is exactly what I need. How do I go about equipping my team to do this?
Yeah. So if it's the Chief Data Analytics Team Officer, somebody on the analytics side, I think the key thing is to potentially get training, right? I think this is a key skill that a lot of analytics professionals, data professionals need. And with some training, they can get the foundational pieces in there. I think also in some cases, some coaching may be necessary to really apply the principles to the day to day kind of examples that they're running into. And then I think it's about getting kind of some opportunities, opening up some opportunities for people to really tell data stories. This is kind of a new concept. And so one example that I was talking to a chief analytics officer and he was looking at a situation where they have this regular report that they would deliver to the organization once a month.
And it was like two hours long or something like that. And they would just go through all different aspects of the business and it had grown in popularity throughout the company to the point where they had a number of leaders coming to that meeting. And obviously reporting is great. And that's great exposure for the analytics team, but I challenged him to see if he couldn't carve out some of that time to focus on telling a specific data story. Now it's not like you're going to say to people, Hey, we're going to... Now we're going to shift, you know, let's do the reporting. Let's shift over to telling a data story. What you're going to do is we're going to do a deep dive, right. So we're going to, yeah, we're going to cover, let's say you have an hour. Instead of using that whole hour to tell or share the findings of a report, we're going to reduce that down.
Can we cut it down to 40 minutes? So we have 20 minutes or we have, maybe we can cut it down to half an hour. We have another half an hour to do a deep dive. And the deep dive is really going to dive deep into a particular area, share of specific insight. And now you have an opportunity to really educate them much more deeply on a particular aspect. Maybe, Hey, we found this really interesting customer segment or there's this process that needs to be reexamined, or we have this change in the industry that we're seeing. And then just doing the deep dive on that. And so I think as there's this cadence of starting to share data stories, I think you're going to start to see people. Oh, wow. That's really interesting. We want to learn more? Do you have other stories that you could tell us and it'll make different teams and managers curious and then your analytics team will have more opportunities to work with them, help them flush out these potential insights and tell more data stories.
So I think it's part of it is there is a shift in mindset that's going to have to occur, shift away from just doing reporting and introducing opportunities to tell data stories, but you will have to nurture those opportunities and obviously nurture the skills of the employees so that they can deliver on data storytelling as well.
Does the story you're telling change based on your role in the team?
I don't think the story changes. I think it's more based on the audience. And so if the audience changes, that's where you're going to be tailoring that data story to each audience differently. You may have the same insight. Now, as I think about your question, maybe if I'm the data engineer, or if I'm the data analyst or the data scientist, maybe there might be credibility things that go on with, why is this person telling me or sharing this insight? Why does it matter that... If it came from the data engineer, is there a different perception around that insight as opposed to if it came from the data scientist or the data analyst or somebody else or the manager. So I think that would be the only thing I would think about. Is there any kind of credibility issue that might influence or shape how people perceive the data storytelling coming from particular role, but I don't think that would be a big factor. I think it's more about the insight and making sure the insight is tailored to each audience.
And that might be if I'm presenting it to a technical audience, as opposed to like a business centric audience, that's going to change how I deliver that story. It's going to change the detail that I go into and what I emphasize with each of those audiences.
You mentioned the importance of training. Is the importance of training around a missing skill. And if so, what are the most common missing skills?
Yeah, I would say the most difficult part for a lot of analytics professionals. So usually when I have a workshop or we have a poll, I ask people which of the three elements is the hardest for you. And so basically almost every time, about two thirds of people say the narrative part is the hardest for them. So they get the data, they get the visuals or the visualization aspects, but it's really about how do I pick and choose what goes into the story? And then how do I make sure that it's concise and delivers something impactful to the audience? So I think a key thing that a lot of analytics professionals need to be trained up on is the narrative elements. And how do I focus and tell that story in a concise and impactful way.
Now, as you think about what needs to be done, is this a matter of technology as well? What are the tools that are necessary?
Yeah. In terms of tools, I mean, I think there's certain tools that we need to find insights, right? So I kind of view you don't have a data story if you don't have an insight. And so if you don't have the ability to find insights within the data, that's the first place, right? We've got to find those insights. And that can be as simple as a spreadsheet. You can use Excel to find insights. It doesn't have to be something super complicated, but definitely there are lots of data expiration tools out there, Tableau, Power BI, Looker. There's lots of tools out there that can really help you to pinpoint or identify potential insights. And then when it comes to actually explaining or telling the actual data story, again, I think you're going to need a data visualization tool to create those data scenes that I mentioned earlier. And then, I kind of revert to PowerPoint as being my main de facto delivery tool for crafting data stories often because there may be some limitations in the data visualization tool as well, where I have to bring it into PowerPoint.
And then in PowerPoint, I can add annotations. I can change the titles and do different things to highlight different content in those visualizations. And then often that combination of a data visualization tool with PowerPoint is usually the best way of communicating and building a data story.
So you briefly just touched on those visualizations that you need to start doing. So your best practice or part of your best practice was to bring those into PowerPoint. What are your other visual and explanatory best practices?
Yeah. In my book, I talk about seven principles that I think are important for visual storytelling. So I think the first thing is to identify whether we have the right data and that means seem, well, if found an insight, I have the right data. What I mean by that is we're often transitioning from an exploratory visualization to an explanatory visualization. And that may mean that we need to change the data that we're presenting, because maybe what helped us to find the insight isn't the best metric or data that we need to convey that key point to the audience. And so that's the first thing. Second thing is obviously identifying the right data visualization. There's lots of chart choosers out there and different tools that can help you on that front. But then I think the third point is, okay, yeah, we've chosen a chart that works, but have we configured it to our message in an appropriate manner?
And so an example of that might be where we have a bar chart and we're breaking it down by a couple of maybe regions by product type. And depending on how we're communicating this information, maybe we need to break it down, not by region first, but by product type and then region. Because we're presenting to a bunch of product managers and they don't care about specific regions and looking up and down through all these different regions. They just want to see, okay, what is my product doing across these different regions? And so we have to think about what comparisons we're asking the audience to make. And so configuring our charts is important. The fourth thing is we want to remove noise. And so there's lots of things that we can do on that front, from removing chart junk. If you're familiar with Edward Tufte talked about a lot of chart junk, and that the extraneous load that sometimes by default gets included in different charts, by different chart vendors.
So we want to remove that. Maybe we brought in, maybe we analyzed a lot of data. Maybe we looked at 30 different store locations, and we found that there's a problem with five of them. And maybe when we're telling our data story, we don't need to cover all 30. Maybe we strip it down to just the top 10 with the five that we're worried about, or maybe top 15 or something. If there's any way that we could kind of remove some of that data that really isn't essential to our data story, we want to do that. The next thing, the fifth point is focusing attention. And so this is really about highlighting in the chart where we want people's attention to be focused on. So color becomes obviously a key tool that we use in our data storytelling toolbox. So if we had five different categories that we're looking at, and we want to focus on one particular category, may we use a bold color to highlight that one category and then push everything else to gray scale.
And so that's going to draw attention to that one category that we're talking about. The next, the sixth principle that I talk about is making your data approachable. And there's several ways that we can do that. One of the common ways is just that we don't want to make it a lot of mental effort to process our information. So there's little things that we can do. For example, if we had long labels on our column chart and so they're very long. And so what does Excel and the other tools do? They'll put them slanted so that they can still be read, but reading at an angle or reading at any angle, the content rather than in a horizontal manner is harder. So maybe what we do is we change the orientation. We shift from a column chart to a horizontal bar chart, and now the labels can be read very easily in a horizontal manner.
And then the last thing is to instill trust. And I think there's several ways that we can instill trust that could be simple things like making sure that we highlight the source. Simple things like not truncating a vertical bar access, little things like that can inadvertently make people question whether we're trying to trick them or deceive them with our numbers when that was never our intent. In some cases, the tools were, will default to that. And so we have to watch out for those things and make sure that we're conveying trust and building the confidence of our audience. And so those are seven principles that I talk about in my book.
Okay. That's a really long winded way to say, use more pie charts, but we can just wrap that up as saying, use more pie charts, right.
Load up on pie charts. Yeah. I'm not anti pie chart. I know there are people out there that are anti-pie charts. I just view it as another tool in our toolbox. So when it's appropriate, use a pie chart when it's not appropriate, definitely do not use it.
As you were talking about those visualizations, what popped in my mind was some of the visualizations from the Apple keynotes. I was thinking about the one when iPhone was not the biggest phone on the market, and they created this pie chart that made it seem like it was totally dominating the market. But they showed it in such a way with the... If you're remembering you remember that one, it was a 3D...
... Pie chart. Yeah. And so they skewed it. So it looked like it was huge when it really, really wasn't. Is that part of data storytelling or not?
Yeah, that is not recommended. You're just not going to get very far as a data story. We're trying to build confidence with our audience, right. And not trying to deceive them or trick them in any way. And I think the long term goal is for them to be a trusted advisor. And so when we do things like that, I think it threatens our credibility. And that's not a practice that I recommend in my book or in my workshops.
Well, then tell us, what's your view of dashboards?
Dashboards, I maybe have a different view of them. I know sometimes people think of dashboards as data storytelling tools, and I don't feel like they actually tell data stories. I think what they do is they help us define potential data stories. So I talk about in my book, something I call the insight funnel. And so the insight funnel at the top of the funnel, when we build the dashboard, we start to frame out the story. We don't know at that point necessarily what the stories will be, but we know which key metrics are important to the business. Because we know what the business objectives are, what the business goals that the different teams are working on. And so that then helps us to choose what metrics go into our dashboards and how we build those dashboards. And so we start to frame the story.
They're going to look at these windows into the business, so to speak and start to identify potential stories. They'll make observations, they'll see an anomaly, they'll see trends or patterns in the numbers. And that may lead to them, depending on the type of dashboard, where maybe they can actually do analysis in that dashboard and dig in and discover what the insight is. Or it may spark something where they're like, oh, we've got to take a closer look at this. Let's go analyze it in some analytics tools and come back with an answer. So I kind of view the top of the funnel of this insight funnel. That's where the dashboards are. And then once we have an insight, then we need to craft the data story around that insight. And so that may mean, in terms of a dashboard, you're going to have all kinds of metrics and dimensions represented there.
Not all of those are going to be relevant to the data story we're telling. So we would essentially want to strip those out and just be really focused, really selective with what we're focusing on for our data story. So that's kind of the difference between what I feel the role of a dashboard is helping business people to ask questions, to make observations of the data, spot things in the numbers that stand out and that need to be addressed. And then, once we do some further analysis and we have an insight to share with others, that's where the data story comes into play.
So in my experience or I would say, in my opinion, I think it's the other way around. I think the dashboard is the manifestation of a data story. That you've created a data story and you've identified some insight and you want to keep looking for that insight. And does that insight keep on happening? What's your thoughts there?
Yeah. I see a data story. I mean, for a data story, we need it to be fixed. We need to be static. We need to cap. It's almost like you're using snapshots of the data to tell the data story. The problem with the dashboard is if it keeps changing, we can obscure or miss the key points that really help to tell a specific story. And so an example of this was recently when I was at Domo, we had rolled out some new feature called stories, which was basically just a way of laying out our dashboards in a specific way with some annotations. And as we were going to prepare for our conference, our breakout session, the data that we had where we were telling a specific story had moved out of the frame. So basically, the anomaly that we had in our data and all the other things that were connected had moved out of the window.
And so, because it's like a CCTV camera. Sometimes you're going to have anomalies that occur and they kind of pop up in a dashboard that's dynamic and rolling. But in the case of a data story, we want to have static snapshots of those key moments and tell that story. And if it constantly is refreshing and updating, we may lose sight of the story we're trying to tell, or maybe a different story starts to emerge. But again, if we haven't crafted everything around that particular news story, it may be incomplete in terms of the information we're showing. So I really do look at it like we need to take snapshots of the data. And then that's what goes into telling the story. If everything is moving and changing, that means that it's going to be very hard to tell a specific story with a dynamic kind of arrangement that a dashboard provides typically.
And you've spent most of your adult life in Salt Lake City. Do you think that's affected your view on analytics as well?
Yeah. I mean, one of the interesting things about Utah, I don't know how many people know this, but there's actually a very vibrant analytics community here. And it came from, it was basically started by when I was working at Omniture. Omniture was acquired by Adobe in 2009. And then they constructed a massive building here in Utah, in Utah county. And so there, you had this, now this mothership kind of where from that a lot of other companies, data related companies, analyst companies like Qualtrics, Domo was another one. And then there's just a multitude of analytics and data related companies here in Utah now. So it's like a kind of a, we call it Silicon Slopes, kind of modeled after Silicon Valley, but it's really been a great experience. I've worked with a lot of really, really amazing people, both from Utah originally, maybe they grew up here in Utah, but a lot of transplants as well.
And so there's a really robust and great community here. And so that's been something that I think has really helped my career working here in Utah and just seeing all the innovation and the community here around data and analytics.
Yeah. I'd have to agree with you on that. I know a few of the Utah, or I should say Salt Lake City Data Glitterati and they're all very nice people, very sharp people. So you're in good company there. So those of you who aren't watching this on video, you probably won't see the two comic books that Brent has right behind him. To start off with, what are the two comic books that you have framed behind you?
Yeah, one is Incredible Hulk 181. Now, if you're a comic geek, you'll know that's the first appearance of Wolverine, a mutant, very popular character. Probably one of the most popular Marvel characters. And so I actually have that comic book. I got that, a copy of that when I was a teenager. And then the other book I have here is an Amazing Spider-Man cover for, I think it's number 40 where he's battling the Green Goblin, and I don't have that book, but it's on my want list. And so I have a number of other amazing Spider-Man comic books, incredible whole comic books. I really like the Marvel Universe.
And are those the only two posters that you have?
I have other ones around. Yeah, I have Amazing Spider-man. I have Amazing Fantasy, which is the first appearance of Spider-man. And then I have X-Men 100 and I have.... Yeah, I'm kind of a geek. So I'm going to have a lot of comic books here in my office and it's kind of, it ties into visual storytelling, right. So if we think about a comic book, it's the combination of both visuals in terms of the panels that are created or sketched art. And then also words. Words are used a lot in comic books. And so that combination really maybe ties into my whole focus on data storytelling and how it's essentially this visual storytelling like comic books are.
So go deeper. Tell us why you think that a comic book would be a good source for our visual storytelling?
Yeah. I mean, if we think about a comic book, which is different than just a regular book, right. Which is a series of words, right? So it's just text in a book and I love books too. But comic books really bring something into this where they combine both the visual and so the scenes or the panels of a comic book, they bring in and tell what's happening and the emotions and the interactions of the character and the scenes. And like, so we can see, and we can envision what's going on. And then you have the word bubbles and different things that help to bring in the dialogue and narration from the narrator of what's going on and what's happening. And I think that combination of visuals and narration is really what we're doing with a lot of our data storytelling, both in terms of the annotations, obviously that's the clear thing. but then that structure that happens behind the scenes. There's a certain structure to storytelling where we have the, as I mentioned earlier, the hook and we have the aha moment.
So we have the climax and the inciting incident and the different action sequences that occur. All of that coming together in a comic book is a lot like what we have to do with our findings from our analysis. We need to pick and choose what goes in and how do we best represent that information visually? So those are all considerations that we take when we're building a data story. And I think it kind of mimics what comic book writers and artists are doing with comic books.
Have you ever put the word ‘POW’ into one of your data stories?
I'm trying to think. I don't think so, but I would be, I'm highly tempted. If you've read any of my books, I'm kind of a nerd and a geek. So I bring in Indiana Jones, you'll see that mentioned in my books and a lot of Star Wars references in addition to comic book references.
What data story would you like to put into a comic?
I think what would be neat would be to really do something meaningful. You know, there's lots of data out there. Obviously, we have a lot of issues out there with gun control, gun laws. For me, I think that would be something that I would love to look into that topic and really dive into it and see if we can't present that information in a way in which it'll actually maybe connect with people and kind of show them some of the issues that are going on and how we can maybe overcome those challenges. But that might be something where if we can take these societal kinds of issues and portray them or convey them in a way that's meaningful and impactful, I think there could be some good there. And obviously, Hans Rosling with what he did with a lot of his data stories, they were very focused on societal problems and issues. And I think there's some good that can be done with data storytelling there.
Yeah. As you were talking about it, one comic book that came to mind was the Superman one where instead of Superman having been raised in the US, he was raised in Soviet Russia.
Do you remember that one? Have you read that one?
Vaguely. Yeah. I kind of remember that, but yeah, very interesting.
It was very interesting because-
It's very similar to the, what if scenario? So Marvel has a lot of, what if. What if Spider-Man was the punisher, what if all these kind of crazy ideas and even, yeah. I mean, some of those.... They did a whole series too on Disney Plus on what if scenarios so very interesting.
So what are you finding interesting in analytics right now other than data storytelling?
I would say the potential for artificial intelligence to really augment our ability to find and tell powerful data stories. I don't think that the technology is there yet to replace us as data storytellers because as human beings, we can process the context a little bit better. We understand our audiences a little bit better. But I think there are opportunities for technology to augment our ability to find insights, especially, I think because right now, obviously we have a lot of data that we're having to go through and often it can be really time consuming if it's a manual kind of labor intensive approach that we're using. And so if we can use the technology strengths where it can crunch a lot of numbers at the same time and look for certain things, I think that any kind of way that the technology can help us on that front is really exciting and can help us find insights faster and maybe better insights. And then we can tell these amazing stories with them. So for me, that's exciting.
Could you describe another challenge that you've seen frequently that isn't related to data storytelling?
I would say another challenge that is not related to data storytelling is just, how do you stay aligned with the business? I mean, my analytics consulting career, I think that has been one of the key falling points that the data team or the analytics team is not connected, closely connected with the business goals, with the business strategy of what's going on. And whenever that happens, whenever you have a gap between what they're working on, and what's important to the key stakeholders within that business, all of a sudden, you're introducing opportunities for business leaders to get frustrated where they're just not getting the information they need to make decisions because it's based on old information or an old understanding of what they needed, but their goals have shifted. Their priorities have shifted and now they don't have the information they need to make decisions.
And so I think that's a critical part of managing a data team, making sure that you're aligned with the business goals and priorities of the internal customers that you have. And so any way that you can form a good working relationship, make sure that you're always on top of shifts in those priorities. And they do shift, they shift regularly. You get a new leader of a team, you could see entirely different priorities come into play. And so being agile and being able to adapt to those needs of the business is really important. And sometimes that's where I've seen teams fail.
Okay. What do you never compromise on?
Never compromise on values or integrity. I mean, I think those are important things that we need to protect. And obviously we could be, we can tell lies with our data stories and if that's our goal or objective, then you know that's not something that I want to do. I'd never compromise on that. I want to be honest with my audiences. When I tell a data story, I want to bring the data, present it in a way that's honest and truthful and I'm not trying to deceive or twist the data story to my narrative or my agenda that I might have. So I think it's important to be ethical and preserve your values and your integrity when you're data storytelling.
And we haven't talked about it, but how would you keep your biases in check as well?
I think that's something we need to make sure we're aware of what our own biases are because they're going to shape, going all the way back to our analysis, right? It's going to, it's going to shape the data that we pick and choose to analyze. And maybe we have a hunch or a preferred outcome that we're trying to drive towards. I think it's important to be mindful of what we have there, make sure that we're aware of our own biases. Maybe we have people challenge our data story before we share it, because it might be easier for them to spot a bias that we have that maybe we don't even realize we had and it affected our storytelling or affected our analysis. So I think the key thing is trying to understand what biases we might have right from the beginning.
Brent, thank you so much for being on the show. I really appreciate it. You've shared a very unique viewpoint and thanks for letting me look at your comic books the entire time.
Thanks, Jessie, for having me.
Another great story, another perspective shared on data, and the tools, technologies, methodologies, and people that use it every day. I loved it. It was informative, refreshing, and just the right dose of inspiration. Remember to check dreamteam.soda.io for additional resources and more great episodes. We’ll meet you back here soon at the Soda Podcast.