Ep 118 | How to Drive Change using Data Storytelling with Brent Dykes
This week Brent Dykes, author of “Effective Data Storytelling: How to Drive Change with Effective Data, Narrative and Visuals,” joins Allison Hartsoe in the Accelerator to talk about driving change with data narratives. If you have ever wanted to know why your charts and graphs fell flat and what you can do to keep your audience riveted, then Brent has the answers for you in this episode. You’ll learn why a data story is not a dashboard, and how a data story should flow to achieve excitement.
Please help us spread the word about building your business’ customer equity through effective customer analytics. Rate and review the podcast on Apple Podcast, Stitcher, Google Play, Alexa’s TuneIn, iHeartRadio or Spotify. And do tell us what you think by writing Allison at info@ambitiondata.com or ambitiondata.com. Thanks for listening! Tell a friend!
Read Full Transcript
Allison Hartsoe: 00:00
This is the Customer Equity Accelerator. If you are a marketing executive who wants to deliver bottom-line impact by identifying and connecting with revenue-generating customers, then this is the show for you. I’m your host, Allison Hartsoe, CEO of Ambition Data. Every other week I bring you the leaders behind the customer-centric revolution who share their expert advice. Are you ready to accelerate? Then let’s go! Welcome everyone. Today’s show is about how to drive change with data, and to help me discuss this topic is Brent Dykes. Brent is the author of the book, Effective Data Storytelling, how to drive change with effective data, narrative, and visuals. Brett, welcome to the show.
Brent Dykes: 00:45
Thank you, Allison. Great to be here,
Allison Hartsoe: 00:47
Obviously, this isn’t a book you just kind of wake up and write, tell us a little bit about your background and how you came to write about this topic.
Brent Dykes: 00:54
So, my background is in marketing, even though I’ve worked in analytics. Most of my career, I’m actually a marketer at heart. And one of the things that I was very skilled at early on in my career was PowerPoint. And so, I did a lot of presentation building, other people’s decks and different things. And then eventually actually started a blog post and published over a hundred articles on PowerPoint, ninja.com, still up there. I haven’t contributed to it and several years now, but there’s still a hundred articles up there, and it still gets a fair amount of traffic. And then, as I started working in an analytics, I started to find that there was a real problem with communicating insights and data. And so one year, it was back in 2013, I realized that data storytelling, that combination of combining the presenting of information with the analytics data felt like it was kind of the sweet spot for me personally, because I could take all of the presentation and design skills that I had developed and then combine it with what I was really doing as a day job consulting and working in analyzing data for different customers when I was at Adobe.
Brent Dykes: 01:57
And so, I pitched the concept of doing a data storytelling breakout session at an Adobe summit. And I had to kind of do a hard sell, but I was able to convince the product marketing guy that it needed to happen. And it ended up being a smash hit, lots of people enjoyed it. And that was kinda my first signal that I was onto something here. And then, from there, I would present each year at Adobe. And then when I joined almost same thing and I wrote articles on Forbes, and one of my most popular article probably generated more than 250,000 page views was an article that I wrote on data storytelling, being an essential skill that we all need to have in today’s data economy. So again, that was another signal point. And then, the last point was when I would present at conferences about data storytelling. I would have people come up to me afterwards and say, do you have a book, or do you have something on this? And so it was kind of like just all those kinds of signals indicated to me that I needed to write this book. And so that’s kind of how I got into data storytelling.
Allison Hartsoe: 02:56
I feel like that is such a great theme because every single place I turn, there are cries to tell a story with data or to provide actionable insights. And my original thinking on this was that this is just people who were desperate for insights because they’re getting onslaughts of data, dumps masquerading as reports. And I think that need remains strong, but really we have a lot of analysts who are working so hard to find great insights. And then they come running out with their insights and people say, Oh, well it needs to be a story, or I don’t understand it, or can it just be more actionable? Why are these analysts failing to get traction?
Brent Dykes: 03:38
Yeah, that’s a great question. And I’ve had many people reach out to me after I published the book. And as I was working on the book to say, I’m comfortable with the data. I can find insights, but where I struggle is on that communication side. And so a lot of that comes down to, there’s a number of techniques that we can apply, obviously, understanding your audience and having empathy for your audience. And then, I look at data. Storytelling is having three key elements. And so one being obviously the data. So you need the data. Then you have the visuals, which is when we think about data storytelling often, that’s what everybody thinks of its data visualization. And then the third piece is really the narrative. And when we think about narrative, I think of more than just texts. Sometimes people associate, Oh, you’ve got some text with your maybe it’s annotations or whatever.
Brent Dykes: 04:26
No, it’s more than that. I mean, obviously, you’re helping to explain the information to the audience. That’s a key reason why we have narrative, but it’s also the flow or the structure of how we’re introducing the insights and information that we’ve uncovered and how we explain it in a linear fashion to the audience and almost taking them through the story arc. If we think of traditional literary works, they have an arc that they follow with a climax and kind of the setting and set up at the beginning and then maybe an inciting incident that occurs that gets people’s interest. And then you have that rising action. So I’ve taken that and applied it to how we can share insights. And I think that’s one of the key things that’s going to really help analysts and other data-savvy people to really, okay, this is how we translate insights into something that’s going to resonate and engage an audience.
Allison Hartsoe: 05:14
So, I want to underscore something that you said to make sure the audience really understands that data storytelling is not just putting text against data or having pretty visuals, that there is a key psychological component of the narrative. Can you talk a little bit more about what’s happening under the covers when you tell a story versus when you just present data?
Brent Dykes: 05:37
Yeah, absolutely. So one of the things that I didn’t realize when I was earlier in my career is that I felt like if I had the logic, I had the reason I had the data, essentially I could bring that into a presentation, share that insight. And then any decision-maker who’s using logic and reason to make decisions should able to see the value of that insight and then elect to put it into action. Now, what I didn’t realize was the impact of emotion and how emotion is a big factor in decision making. And actually, one of the people that I found that did a lot of research into this area was a USC professor by the name of Antonio Damasio. And he actually worked with people that had trauma to their brain in the sense of their emotional centers and were almost like Vulcans. And if you don’t know what a Vulcan is, kind of a star Trek reference there, but it’s somebody who basically has no emotions and operates in that way.
Brent Dykes: 06:31
So, in this case here, they’re very logical and their reasoning and the, and to your eye, they would look like normal human beings. However, what he found when he worked with these patients is that they had a lot of trouble making simple decisions. And so an example of this would be, he would tell them, Hey, let’s go to lunch this Tuesday, where would you like to go to lunch? And then the decision that you or I could make in maybe a minute or two would literally take these individuals 20 minutes, 30 minutes to kind of come to a decision cause they vacillate back and forth between the different options. They say, well, if we go to the sushi place, they have the special on Tuesdays. So that would be really good. But then wait a second. The Italian place is probably easier for parking. And then going back to the sushi place, I really liked the servers there, you know, and they vacillate back and forth and really struggled to make a quick decision.
Brent Dykes: 07:17
And so, emotion is actually really what helps us to come to quick decisions. And if you’ve read the book, Thinking Fast and Slow, by Daniel Kahneman, he talks about two systems. And so we have system one and system two and system two is when we think about the brain processing and analyzing information, that’s what we think of system two, where it’s a conscious effort where we’re reasoning through the numbers. But a lot of the information passes through system one and system one is really where our intuition and our heuristics and pattern-seeking kind of abilities come in. And also, our cognitive biases exist in this system one, cause it’s trying to process a lot of stimuli and make sense of it and then pass that to system two, to kind of as needed, make a decision or evaluate something. But a lot of our reasoning is actually happening unconsciously in system one.
Brent Dykes: 08:09
And so, what that does is sometimes these cognitive biases, and also emotion comes into the decision-making process through that system one. And so that becomes a challenge when we’re only coming with data. And so one of the advantages of the narrative and there’s a lot of research into the unique effects that narrative have on our brain in the sense that if one of the research that they did was looking at us when we’re just listening to statistics or facts, there’s a couple of regions in our brain that will light up. And when these brain scans that they do, and it’s just purely, those two areas are purely just associated with processing words and language. And so that’s it, but with a story, when they looked at people who were listening to stories, all of a sudden they saw multiple regions of the brain light up in the that they were feeling, not just hearing the statistics, but actually feeling what the other person was sharing with them.
Brent Dykes: 09:01
And then there are other studies where Yuri Hassan, he has got a lot of good tech Ted talks. He looked at how the brain looking at the brain waves of the storyteller and then the brainwaves of the subjects who are listening to the story. He noticed that there was almost a connection where the brainwaves that the two, the storyteller and the listener were mimicking each other. And so, they were experiencing what the storyteller was sharing and almost coming in sync if you will. So, there’s a really strong connection that we can get through sharing stories. And I kind of view it as that bridge. If we look at our data and our insights that we have, how do we bridge to that emotional part of the brain? And I think that our gateway to that or our bridge to that side is the narrative and the emotional kind of connection that it creates with the audience that a narrative structure and maybe even humanizing the data that we have for the audience forms a connection with the audience that makes our insights resonate with them, become clear and become more memorable.
Allison Hartsoe: 10:01
I guess if we use an analogy, it’s like when you present data, you’ve got a fortress with a gate, and you can either take the easy path by tunneling under and going right to the emotion that lets you through the gate. Or you can run right smack into the gate and pound on it with a log and try to get through. And I think what most analysts are doing is they’re bringing the log and the logic heavy into the gate over and over and over again. And people are like, I don’t want to listen to you. I’ve got other things to do. But when you make that emotional connection or that story, you bypass the gate. You end up with, I think, the concept you said in your book was neural coupling, which came from the Hasson research. You start to, it’s almost, I think about it like a choir when people go to church, and they listened to the choir, everyone who’s singing in the choir, their heartbeats all get in the line, and it’s similar. And that you’re aligning with your audience almost like that Vulcan mind meld, you know, you’re connecting.
Brent Dykes: 11:01
Yeah. And actually, the interesting thing, the way we typically react to data and facts is we go on guard. We put up our defenses. So, your fortress analogy is very correct because we raised the Drawbridge. We were on defense. We don’t want to be tricked or deceived by this data that we’re looking at. However, when neuroscientists looked at how people respond to stories, it’s a much different reaction. They’re actually less likely to nitpick on the details. They’re actually, they want to see where the story’s going. And so, their defenses come down, and they’re more open to hearing perhaps new ideas and new perspectives that they wouldn’t have considered if you just came with the data. And so, it is some people have called it like a Trojan horse to get into the fortress. And I don’t really like that angle cause, I don’t think the goal really is to deceive or to manipulate people.
Brent Dykes: 11:52
And the example, the analogy I use in the book is that if you think about the brain being all these pathways, and it’s almost like a metropolitan where you have a metropolis where you have all these highways and different things, and what the stories do is they take advantage of the express lanes, the HOV lanes, where there are these natural pathways that are opened up to stories where they can travel faster, they can avoid the congestion, the traffic congestion of trying to process data and insights. And almost they can take advantage of these expressways within the brain to get information to the audience at a much easier and faster way than just having them perhaps nitpick the details and really scrutinize everything. So, it’s, I like to think of it that way as a positive, that we’re able to transmit our information more easily and faster to people by leveraging the narrative.
Allison Hartsoe: 12:41
So now I was trained in journalism, and when we think about story, we oftentimes think about set the hook and then drive into the facts afterward. But when you are talking about storytelling, you’re not really talking about the same thing. I think you’re talking about what you call story framing. Can you talk a little bit more about that angle and what it really means to structure a story correctly?
Brent Dykes: 13:05
Yeah. So, there’s a couple of things there. If you think of journalism and you can push back on me, there’s this thing with leading with the lead, right? You have this lead. And so, what is that that’s often called the inverted pyramid. And so, you have the most important information will be at the very beginning of an article, the first one or two paragraphs. And so essentially, you’re giving away what would be the climax in a literary story at the beginning.
Allison Hartsoe: 13:30
Give us an example of what that would sound like.
Brent Dykes: 13:32
Would you want me to use like a data storytelling context?
Allison Hartsoe: 13:35
Any, like a famous movie.
Brent Dykes: 13:37
Let’s go with Harry Potter. How many people have seen Harry Potter? Probably everybody. So, the way that the story arc would work is when we’re introduced to Harry Potter, let’s take the first movie or the first book. We meet this poor kid.
Brent Dykes: 13:48
He’s an orphan, he’s living under the stairs. He’s maltreated by all of his closest relatives and pretty much leading a very miserable life. And so that’s kind of the setting, right? So that’s how we set up something. And then there’s this what they call the inciting incident, and that’s it. Or you refer to it as a hook. And that’s how I kind of call it in my book, a hook, meaning that something interesting happened to Harry Potter. There’s probably some debate about what that actual event was, but in my mind, it’s when he goes to the zoo with his family, and he ends up talking to a snake, that’s kinda like the first weird kind of interesting event that happens to him and kind of shows that, Hey, wait a second, I’ve got some special abilities here. I can talk to snakes. And that kind of starts them on this path now.
Brent Dykes: 14:31 So that’s the inciting incident that starts the journey. And then you have what they call rising action. And so, from there, he obviously gets rescued from his relatives by Haggard and then takes it to Hogwarts. And then he has all of his experience there. And then the climax of the story is where he’s battling Voldemort. And then the resolution of that, where basically the Dumbledore gives his team, the Quidditch cup. And then he returns a hero to his old family that he had. But now he has a new family, a new identity and everything. And so, we can take that same model and apply it to data storytelling in the sense that we have that setting. So, what is the setting for your data story? Well, you’re giving context, right? You’re giving some information on, Hey, we’re going to be talking about our marketing campaigns, or we’re going to be talking about this customer segment or whatever it is.
Brent Dykes: 15:18
We’re kind of setting up kind of what is the status quo, what is normally occurring? And then what we do then with the hook in a case of a data story, that’s where we see something in the data that has increased or decreased, you know, a metric that is spiked, maybe it’s revenue or conversion rate that’s decreased. And so, then that starts our journey. That’s going to peak the audience’s curiosity. Oh, what caused that? What contributed to that increase or drop in a particular metric? And then that’s where we start to layout the information and start to share. Okay. So, this is what contributed to the conversion rate drop. And then, which was also contributed by this. And depending on, you know, some cases, it might be very complex cause there’s lots of information that we’re looking at and analyzing, and that we need to provide context and insight to the audience.
Brent Dykes: 16:04
In some cases, it may be very quick. We could jump from the hook, right to our, what I call aha moment, which is the climax of your data story. It’s the one main point that you want to make to your audience. If they leave your data story with nothing else, you want them to remember your aha moment, which is your main central insight. And that is important because one of the challenges that we have in analytics as we do the data dump, right? So, we, I found this, it was really interesting. I found this. That was really interesting and Oh, and here’s another thing. And here’s another thing. And the audience is left with where are you going? Where are you taking me? Where is the aha moment? Where is the big reveal? Where’s the big insight that I need to take away from this conversation or this presentation you’re delivering to me.
Brent Dykes: 16:47
And that’s the importance of the aha. And I think that’s lost a lot of the time on analysts and people that are very data-savvy is that we have all these different insights and we, unless they’re kind of a set in the sense that they’re related and they are revealing one major insight or when major takeaway, I would generally say to people, you might have more than one data story here. You might want to actually separate them and not tell them all at once. Imagine telling Cinderella and snow white at the same time, we’d never do that. We’d never do that. But we do that all the time in our analytics presentations and our data reports and try and pack in as much information at the same time. And then what happens is we’re like, wait a second. I thought you were talking about the princess.
Brent Dykes: 17:29
Oh, she’s not a princess. She’s actually get mixed up between Cinderella and snow white. So that’s the key thing we need to pick. What is our story? And then they’ll, and then we’re not done when we get to our aha moments, it’s all about driving action and change and value with our data stories. And that means that we need to help the audience to see, okay, that’s great. You’ve now discovered what’s contributed to that decrease in our conversion rate. What do we do about it, Brent? How are you going to show us what we need to do? And that’s where you, as the analyst, need to then think, okay. Yeah, we’ve identified the problem. Here are the three solutions. We can reduce spending in this area and invest more over here or B we can look at a new vendor because clearly, the vendor is contributing to this issue that we’re working with a service provider or three, we could do a combination of maybe the two first options.
Brent Dykes: 18:16
I mean, again, it totally depends on the situation, but providing options, helping guide them with some recommendations and some cases I’ve had people say, well, Brett, I don’t know if I’m in a position to really say what the recommendations are. Well, then you can at least start the conversation, but you get the right people in the room, the right decision-makers, the right key stakeholders, and you’ve guided them to the aha moment. You’ve kind of laid out some potential options. And then what comes out of that discussion may be the actual solution. But here’s the kink. The kink here is that you’ve created a data story that compels people to act that compels people to make a decision to move forward. And that’s the biggest complaint I’ve seen in having worked in analytics for the last 16 years. It’s like, ah, I’m just not seeing the value from our product to the platform or the data, you know, and a lot of the time, you know, I talk about failing, and that last mile and that last mile is you’ve done all the hard work.
Brent Dykes: 19:08
You’ve done a ton of the hard work to collect the right data, to process it, cleanse it, analyze it, visualize it. And now you’re on that final stretch where communication becomes really important that we don’t throw away all of that hard work, all of that effort and time and investment, that’s gone into it. And we just communicated in a clear and meaningful way. So that decision-makers and people that need to make a decision on this data can see what they need to see and then clearly have a path to act on it. And then that closes the mile. We’ve now taken the insights from all that hard work that we’ve put in on the data side to now actually driving action and driving value from all of that effort and investment.
Allison Hartsoe: 19:47
There’s a couple of things I want to circle back to over about what you said first, the concept of finding the aha moment. Can you talk about what people usually do versus the beauty of finding the right aha moment and then structuring around it?
Brent Dykes: 20:02
Yeah, I mean, I really emphasize before you start building a data stories to really see if you have an aha moment. And what I mean by that is do you have that really powerful, actionable, you know, we talk about actionable insights, and you’re going to have your aha moment is going to be an actionable insight. And so that becomes your destination, you know, without, I think that’s what happens a lot of the times where we I’ve even seen it, even in the world of data journalism, where I feel like I’ve started going down, reading an article that is, it’s got the visualizations, it’s got a great narrative, but what it’s missing is like, what is the destination? Where are you taking me? And what is the big aha you know, the big insight, the central insight of this story that you’re sharing with me? And it’s, as I’ve mentioned to people in the past, I’ve said, I was actually, I won’t give the name of the publication that did this data story, but I felt like it was an interesting topic.
Brent Dykes: 20:56
It had really good visualizations. It had a lot of commentary and annotations and a lot of rich information, but it felt like it was just like, here’s something, Oh, and here’s something else, and here’s something else and here’s something else and that’s fine. That’s you can do that in a data story, but is it building up to the aha moment in this case? I felt like there was no aha moment. It was just, here’s some more, here’s some more, here’s some more.
Allison Hartsoe: 21:21
You just wandering around in the woods.
Brent Dykes: 21:22
Yeah. Here’s some more, here’s some more, and then you get to the end, and you’re like, okay, but what did you want me to take away from all of these insights you just gave?
Allison Hartsoe: 21:29
Yeah. I like the phrase. Um, and because of that, you know, which I think they use in the Pixar framing, which is something happened here’s the steady-state. And because of that, you start to reveal the different pieces and how they fit together until you get to that aha moment. I always liked that, um, question when you’re trying to figure out, does this actually belong here?
Brent Dykes: 21:50
Yeah. In one of my presentations, actually, the most recent presentation that I did actually looked at the Pixar model, and I compared it to my own model. And they actually are very, very complementary in the sense that you can see just, so people on the call know what this is called, Pixar story spine. And it was actually introduced by a teacher by the name of Ken Adams and basically what steps are. And I’ll all kind of tie it back to my model. So, you can kind of see how they work together. But I think it’s in my case, the reason why I share these other models is because I want people to understand this. I want people to grasp the importance of the narrative structure, so let’s go into it. So basically, their first statement is once upon a time there was blank. And then the next statement is everyday blank.
Brent Dykes: 22:33
Okay. So, what is that doing? That’s setting the context, that’s kind of setting up, what are we going to look at?
Allison Hartsoe: 22:38
The normal.
Brent Dykes: 22:38
Yeah, the normal, and then the hook comes when it says, but one day blank. And so that is the hook that starts up the journey. That’s where the character runs into a problem or something happens. And then that changes the trajectory of what normal looks like. And then as you said, because of that and there, and the story spine model, they have three. So, they, because of that, because of that, because of that, and that’s the rising in my model, I call that rising insights. And so, you’re starting to reveal a little bit more, a little bit more, and then the climax comes at until finally blank. And that’s the climax of the story or in my world, the aha moment.
Brent Dykes: 23:17
And then what I’ve seen, a lot of people who’ve written about Pixars model is they stop there. They don’t include a last step, and you’ll see it actually in blog articles and different presentations, they drop the last step and which is and ever since then blank. Okay. And that’s where I go within my model. That’s where you’re talking about next steps and solutions. And for me, because we’re driving value, we’re trying to drive action and change. That’s critical. We can’t just stop at the aha moment. We can’t, we have to keep going and make sure that whatever that new, how are we going to address this problem? How are we going to seize this opportunity? Or how are we going to reduce the risks that we’re seeing, that’s where the magic happens. And so anyway, I bring that up because I’ve found that the story spine and what I call the data, storytelling arc are very aligned and complimentary, and I’ll do a blog post about this in the coming weeks. But yeah, I presented on that, and I’m happy that you brought that up because I think they’re very complimentary and helpful kind of help people understand this concept.
Allison Hartsoe: 24:17
What I think it’s so interesting about this process is as the analysts spend a lot of time finding the insights, extracting the nuggets, and then there’s this additional layer of putting it into the story to make it digestible and understandable and to hit that emotional narrative. But what I find at the very end of the process is you might spend 150 hours to get to a story that lasts maybe. And I’m guessing here 10 to 15 minutes or some smaller amount of time. Is that a good assessment in the ratio of how much time you should spend to figure something out versus the presentation time?
Brent Dykes: 24:54
Yeah, I mean, that kind of leads into something I talk about in the book where I have kind of a two by two matrix that I share, and probably it would be surprising for people to hear from somebody like myself who believes in data, storytelling that do you always need to tell a data story? And I will say, no, no, you don’t. There are situations when you definitely need to tell a data story. And then there are other situations where if you want to tell a data story, you can, but it’s not needed. And the reason why it kind of goes back to your point, that to tell an effective data story, sometimes it can take a lot of effort and a lot of time and you’re right, you know, to deliver that data story may only take 10 or 15 minutes, 20 minutes depending on the audience. But if it’s driving a key decision. So, let me explain what a two by two matrix includes. So basically, one axis, you have the value of the insight. So, you’ve kind of done some assessment of how impactful that insight would be to the business. It could be from a monetary perspective or whatever the size of your business. Obviously, we change that for a big company, maybe a couple hundred thousand dollars, not a big deal, but for us.
Allison Hartsoe: 25:58
Like a decision tree almost. Yeah. If it’s not a very big or high impact insight, maybe you don’t need a story.
Brent Dykes: 26:04
Right? Yeah. So that’s the first criteria. There’s some kind of medium to high. I would say if it’s medium to high, then yeah. You should consider telling it a story. If it’s low valued, then I don’t know if it’s worth the effort. Now on the other axis that I have on this two by two matrix is that’s where I have the type of insight. And I talk about it being a hard insight or an easy insight. And what I mean by that is how easy is it for people to kind of follow or accept? And so, some of the.
Allison Hartsoe: 26:33
Tech sector to execute.
Brent Dykes: 26:34
It could be accept and I’ll get into some examples of what I mean by that also to execute could also be a part of it, but it probably it’s more on the acceptance of it because if we all decide it’s worthy of doing then, Hey, how are difficult that is or.
New Speaker: 26:48
But let me get some of the examples. So if I’m coming back with an insight, that’s going to be unpleasant, meaning let’s take an example of marketing example, the campaign, we just spent a couple of million dollars on for the last three months, and the results are not good. It did not perform very well. So that’s unpleasant information that people do not want to hear. That’s going to be challenging. It might be a disruptive insight. So the way that we’ve traditionally done things on that campaign, we’ve now realized we need to completely overhaul how we’re either running the campaign today and completely pivot to a different strategy or approach, or maybe that insight is completely breaks our business model. So again, if it’s destructive, you’re going to have to spend time data storytelling. Now, if the results are unexpected, that’s going to be challenging for people to accept.
Brent Dykes: 27:33
Or if it’s complex, there’s a lot of complexity involved in explaining the insight. That’s something where you may have to consider doing a data story, other situations may be where it’s risky, and there’s an element of risk to this that needs to be factored in, or it’s costly. And that may be where your execution piece comes in that I’ve got a great insight. We could make $5 million, but we’re going to have to invest a million dollars. And then if you’re a small company then might be, Oh my gosh, a million dollars to get 5 million that’s Ooh, that’s a lot. Or here’s the last one? It’s counterintuitive. You have an insight that runs counter to the intuition of the team that you’re presenting it to.
Allison Hartsoe: 28:08
Oh, that’s a tough one.
Brent Dykes: 28:09
Yeah. So, on the flip side, at the bottom of the what’s easy, and I would say not necessary for you to tell a data story is where the information is agreeable. Hey, that campaign was amazing. Look at the results we had. I mean, everybody will feel good, but do you necessarily need to tell a data story because everybody’s kind of happy by the results.
Allison Hartsoe: 28:28
But maybe you should grease the wheels a little bit by telling happy data stories before you tell an unhappy data story. Just so people get used to taking in your information.
Brent Dykes: 28:38
Yeah. You don’t have to do all the ugly ones, and you know, but yeah. I mean less effort is required for an agreeable story or a conventional story or something if you’re reporting on information, that’s expected like we expected a thousand leads from that campaign and Oh, sure enough, we got a thousand leads. How much effort do you need to put into that data story? Probably not as much effort. If it’s simple or safe or inexpensive to execute or intuitive. Those are examples of areas where we don’t always need to tell data stories, but it does need to happen in those areas where we’re talking about medium to high value. And maybe the insight itself is harder for people to appreciate or to accept or understand.
Allison Hartsoe: 29:18
Yeah. And that’s the right way to do it. So, let’s talk a little bit about the process. So, let’s say that I am sold, and I understand the idea that I should be telling data source stories. I want to try to get a certain aha moment across, and maybe it’s positive, maybe it’s negative, but I want to take action on what I’ve learned. What should I do first, second, third? And maybe along those lines, are there certain gotchas that I should worry about as I try to take action on this process?
Brent Dykes: 29:46
Yeah. I mean, obviously, the foundation of every data story is the data. And so, if you’re confident that you have good reliable information, the data is of good quality. Obviously, that’s going to be a major building block. And so, you want to make sure that your data is, and by one thing is it’s gotta be trustworthy. That’s obvious. So that’s data quality, but it’s also going to be relevant. And that really ties back to your audience, which I would say is another key thing. How well do you understand your audience? What they’re focused on? What are their problems? What are the outcomes they’re trying to drive? What are the strategic initiatives that they have in motion that they care about, that they’re in spending their budget and their resources on, and what KPIs or key metrics or measures are they being bonused on? These are all understandings of the audience.
Brent Dykes: 30:31
And I would say, that’s also another good place. And it’s tied to the data as well, right? Because if you understand the audience and what they’re trying to achieve and what problems they have, et cetera, then that starts to guide you to the right data that you’re going to need. And then, you can determine if you have the right data that’s relevant and trustworthy. From there, obviously, you’re going to be doing some analysis. And here’s the key thing. I think this is where we get screwed up a lot on the analytics side is that there are two phases to any kind of analysis that we’re doing. There’s the exploratory phase. And we’re using visualizations at that phase, but those visualizations are just for us. They need to work for us. They don’t have to be pretty, and they don’t have to be simple or concise. It’s really just whatever helps us to discover the insight that we need.
Brent Dykes: 31:15
And so that’s where we’re exploring the data is we’re drilling in where we’re using these visualizations to help us guide us through this process of finding an insight. And when we find an insight, then that’s where we need to pit to the explanatory version of those visualizations. And I would say that as we’ve gone through and we’ve analyzed the data, we found some key points we need to pause there and look at, okay, so what do we have here? What kind of insights do we have? What is our main insight? Going back to what we talked about earlier, what is our aha moment? Do we have an aha moment? And then what data do I need to support that aha moment? How do I take people to that aha moment? And I like to start with the aha moment. That would be the key thing I want to identify.
Brent Dykes: 31:59
And then what I need to do is jump to the, what is the start of my story? What is that hook? That’s going to get people interested. It’s probably just an observation in the data. And maybe in some cases even say I’m an analyst and an executive comes to me and says, Hey, what’s happening in this dashboard here. We have a jump in our sales what’s causing that. I’d like to understand that. The audience has already given me the hook. It’s that jump in the sales. And so when I come back with my data story, that’s how I kind of kick it off with like, remember when you said you’re typically our sales are here at this level, but you noticed in this last quarter that we had a huge bump in this one category. Okay, well, what I’ve done is I’ve gone in, and I’ve done this analysis.
Brent Dykes: 32:37
And then what we’re doing is we look at all of the data that we have, that we then have to pick and choose, okay, how do I connect the hook up to the aha moment? And really, I don’t want to take them down tangents. I don’t want to lose them in that journey. I want to be very concise and efficient and really only include the additional as I call them rising insights to really connect the dots between the hook and the aha. And then the last step. And maybe this is where some additional analysis is needed is then, okay. So, what, what do we do about this? Aha. And that might be where, because my main focus was just on generating the aha moment and discovering that aha moment. I may not have at that point done some analysis to really say, okay, well, based on this, we should be doing A, B, or C, and there may be additional analysis that goes into exploring what those options could be.
Brent Dykes: 33:26
And so, then what I’ve done is I’ve now gone through, and I’ve started to organize how I’m going to tell my data story. And then the last step is really okay. I have all these visuals that I may have on the exploratory side of helped desk help me to find these insights. And one of the mistakes that analysts will do is they’ll say, well, this visualization that I created speaks to me, it should work for other people. And often, that’s not the case. As an analyst, you’ve spent a lot of time and the tools, you have a lot of time in the data, and there may be a lot of noise in there that you see through. You look past it. You don’t see the noise you don’t see the other additional data points and values in there. But what we need to do is we shift to telling our data story, how do we bring out the insights and the observations in those charts and really communicate to the audience, what they need to hear.
Brent Dykes: 34:16
So, we start saying, we look at the kind of what we’ve structured and then say, what is the ideal chart for this? We may even revisit the chart. We might’ve done some scatterplot or something like that to find something. And then we realized, you know what, that’s too complex. That’s too busy. There’s a lot of noise there. Maybe we just need a bar chart. And so, we may reevaluate, or even the data. We might look at the data and say, okay, I’ve found that we have an increase in sales, but maybe we need to do not only is an increase in sales, but it’s an average, the average order value went up by X percent. And so, we pivot away from, we might be introducing ratios or different calculations or different things to kind of explain our story.
Allison Hartsoe: 34:54
But I think the key concept there is they need to edit and fine-tune before you get to that final version, just like any good story. Most stories are heavily edited and have multiple perspectives on top of them to really drive the music of the story. And I think we can underestimate that. Like there’s a tendency to want to keep it close and then to present. And yet, that’s probably a sure recipe for failure.
Brent Dykes: 35:22
Yeah, no editing is critical. And I talk about three data forgeries in the book and one of the chapters and the data chapter, chapter five, and I talk about three data for, I call them data forgeries, cause they look like data stories, but they’re actually not what I feel are effective data stories. And so, you’ve hit on one, which is, I think this is a mistake that the analysts make is that they go through that process. They use the exploratory visualizations to then communicate their findings to the audience, and they don’t put it in the narrative, but it’s that editing process of thinking, okay, how can I best communicate this insight to an audience? And so that is the first mistake that I’ve seen that’s made by analysts. Another mistake that I’ve seen that’s made more from the business side is they already know the story they want to tell.
Brent Dykes: 36:07
And so, go find me the data points that support my story. And so, then that breaks get, it looks like a data story, but it doesn’t have that rigor on the data because it’s really just cherry-picking the numbers and either.
Allison Hartsoe: 36:20
Tough spot.
Brent Dykes: 36:21
Yeah. Consciously or unconsciously, you’re selecting the information that supports your agenda, your story, or your position. And then you build a data story that way. And then the third data forgery that I talk about in my book is where, and I’ve seen this in situations where you look at a chart or a series of charts that are very interesting. They’re very pretty. They’re very obviously interesting visuals. There’s obviously maybe a designer that’s being involved in creating the visualizations. And then as you start to poke at the charts and maybe interact with them, you start to see, wait a second. What is the main point that this is trying to be made? And you realize there is no main point. It’s almost like, the person created the charts, hoping that somebody would find a story and take meaning from what they’ve created. But again, I feel like that’s not a data story. You have to have the aha moment in mind. You have to build your story around that and then guide the audience. I like to think of that as you’re guiding the audience through the information and helping to enlighten them to a new insight that they wouldn’t have seen without your help.
Allison Hartsoe: 37:19
That’s perfect. So Brent, if people want to reach you or want to get a copy of your book, which is again, effective data storytelling, how to drive change with effective data, narrative and visuals, how can they get in touch with you? How can they get a copy of the book?
Brent Dykes: 37:33 Yeah. So, I have a website for the book. If you want more information on the book, it’s effectivedatastorytelling.com. You can also find the book on Amazon. So just look for effective data storytelling, and it’s right there. And yeah, that, those are the two best ways where you can connect with me on LinkedIn. I’m also, I love to connect with people who are into data storytelling and leveraging data in effective ways. So yeah, always.
Allison Hartsoe: 37:55
Excellent.
Brent Dykes: 37:56
Yeah. Those are the best ways.
Allison Hartsoe: 37:57
Yeah. And I’ve actually read quite a few books on storytelling and various flavors of how to communicate with visuals. And I can say, I highly recommend your book because not only is it effective at helping you understand what to do, but the narrative of the book itself is engaging. And I just really want to compliment you on making an interesting book out of something that might not be an interesting topic for most people. So, this is a great job. So as always, links to everything we talk about, including Brent’s book, are at ambitiondata.com/podcast. Brent, thank you so much for joining us today. It’s been a really enlightening conversation, and I know we’ll get a lot out of the nuggets that you’ve shared with us.
Brent Dykes: 38:41
Thanks, Allison, it’s been great to chat about data storytelling. It’s always great to talk about this topic, and thank you for the opportunity.
Allison Hartsoe: 38:48
Remember everyone, when you use your data effectively, you can build customer equity. It’s not magic, just a very specific journey that you can follow to get results. See you next time on the customer equity accelerator.