Episode 83: Data Visualizations

Episode 83: Data Visualizations

Episode 83: Data Visualizations 560 420 Carlos L Chacon

We’ve all been there – you have some data and you need to make it pretty for the report. With the all the new reporting options you have probably been tempted to find a new way to visually present information.  In this episode of the podcast, we are talking about data visualizations and our guest, Jonathan Stewart, has some issues with the way we currently go about it.  Jonathan has been on the show as a panelist and I have really enjoyed his perspective on things so we invited him on the program to discuss some ways you can go about improving the data visualization of your reports, without adding additional burden on process.

Psychology plays a big role in data visualization. To make a good report, before even starting your tool, you should be able to summarize what the report presents. One of the major aspects of psychology is color and how we perceive it. Jonathan explains how we can use this while visualizing data.  We discuss color blindness and I haven’t heard this discussed too much elsewhere and I was curious to know how we can solve this problem with an easy button.

 Episode Quotes

“One of the major aspects of psychology is just color, right? What colors you’re using? And this goes back to my storyboard and you see everything comes back to storyboard, right because that’s like. Actually I have everything done before I ever open a tool. Even the colors picked out should be done before I ever picked a tool.“

“Don’t think of it as, you know, I have to this extra work. Think of it as you’re doing it to prevent extra work. If you develop a framework, you develop your own storyboard and you get your stuff there“

“Out of 30 people in the room, there’s 3 people at least in that room that are colorblind. And then too, there’s not just one type of colorblindness. There are multiple spectrums.“

“We’re datafying everything. We datafy everything from our footsteps, to our calories, to our heart rate, to traffic on the road, we datafy everything because we like to be able to see and analyze things.“

“Are we using the numbers to prove our bias or we’re using the numbers to disprove our bias and to show the proper thing?“

Listen to Learn

  • How to do a storyboarding before even opening a tool to create a report
  • How to include storyboarding in data visualization process without investing too much time
  • How you can prevent additional time spent on report modifications
  • Color psychology behind data visualizations
  • How to create a report that colorblind person can understand
  • How the quality of the source data affects the end result
  • Predictions how the reports and data will be used in the future

Jonathan on Twitter
Edward Tufte and data visualization
Colour Blind Awareness
See you charts as a color blind person would
Building a data visualization storyboard

About Jonathan Stewart

Data VisualizationJonathan is a SQL Server professional of 18 years. His experience ranges from BI architecture, data modeling, data warehousing, database architecture, database administration and strategic IT planning for Fortune 100 companies. He sees data in everything and is constantly looking for ways to organize, read and analyze whatever data set he can get his hands on. He is passionate about data and making sense of it. He is passionate about using data to effect positive change in his local community of Columbus, Ohio. When not doing data things, you can find him mentoring youth in his community.

Carlos: This is Carlos Chacon.

Steve: I’m Steve Stedman.

Jonathan: I’m Jonathan Stewart.

Carlos: Jonathan, welcome to the program.

Jonathan: Thank you for having me. I appreciate you having me on. It’s a great honor.

Carlos: Yes, always interested at chatting with you. And ultimately our topic of discussion today is data visualizations. Now, this isn’t necessarily a reporting discussion but ultimately I guess the way we present and potentially consume reports or as we say data visualizations. So I guess let’s jump into that. And you have an interesting story why is this an important topic for you.

Jonathan: Well, first off one of the issues that I’ve always seen is that whenever we look at data visualizations first thing we think about is a report, a chart, the tool whether Excel or PowerBI, Tableau, Qlik doesn’t matter.

Carlos: Sure, SSRS, a lot of our people using SSRS.

Jonathan: Yeah, Reporting Services. And that’s how we see it but one of their problems, you know, we end up with problems of was the report right and stuff like that. All those problems we end up having with it can be hedge upfront if we look at it from a different view. I begin to look at things from the focal point, the first point. Look at it from the problem and then begin to look at it as a story versus just a report. And how can we get the story better, how can we tell the story better? So I begin to look at, I’ve always been a psychology junkie, so I always love stuff like how these colors make us feel, how this, you know, things like that. All these things begin to build in me about visualization so I begin to study it more and that became almost my focus. I was already a business intelligence consultant so I was already building reports and stuff like that so I got to see things of how people receive certain reports. You know, why pie charts suck. I got to see all these things as I was building out my career and building my focus. It became my interest. It’s a huge interest to me. Everything from how we perceive the world, color blindness, all kind of stuff like that begin to take shape and I get to see it. And as I learn more, and this is just me in general, as I learn more I want to share more to people. I was like, “Oh, this is amazing. Let me share this with them. This is amazing.” So then it became, you know, somebody was like, “You should just speak about it”, so that’s actually one of the things.

Carlos: There you go. Help us here because we’ve had discussions on this show about many of the different reporting options that are available to us, right? And ultimately while there are different needs that’s why there are so many different tools. Why are we fumbling, on what ways are we fumbling when we actually talk about the presentation layer if you will?

Jonathan: One of the major issues that I see in the problem of visualization is that the current tools are sold as self service BI, self service this, self service pumping your gas, and stuff like, of self servicing. The pumping your gas joke is something that Edward Tufte in one of his blogs he made a comment about. He has a huge issue with people like Qlikview, Tableau, even PowerBI, you know you have our tool you can make this. He’s like it’s not like pumping gas. You can’t just take data and just show it because if it has no context, if you don’t know what it means then it is still is a bad visualization.

Carlos: Got you, so we need to be able to better represent what the data means when we present it.

Jonathan: Right, so one of my focus is that I bring it into, like when I go to a client and they’ve never see things like this before I actually storyboard before even I open up the tool. I’ll go and I‘ll interview. If I get a request, I need this, so I’ll go and I’ll sit and talk with them. And it will be a legit interview, almost like I’m building full blow data warehouse. I’m interviewing businesses and stuffs like that. One of the first questions that I ask is, “What problem are you trying to solve with this report?” So before I even open up the tool, well, one of the things too that we do. When we open up a toll, as soon as we start coding something or writing something whether it’s five minutes, or an hour, or 5 days a week whatever, if it’s wrong we really don’t want to change it because it’s our baby, right? Like we don’t want to change it. We write a certain procedure. I really don’t want to just throw it away. We want to modify it. So even before I get to opening any tool, it doesn’t matter what the tool is, I want to make sure that I understand the problem.

Steve: So one of the things you mentioned there was Tufte and I know, I’ve looked at some of his books and I’ve seen some truly amazing visualizations. And I know one of the disconnects that I’ve had is going from here’s the tools we’re given, SSRS or whatever, or maybe that we’re using to, how do we get from that to some of these amazing things that you’re seeing there? And that’s where; I mean even with PowerBI, there’s a lot more features in there. But to really go to some of the stuff that he shown, it’s a big jump.

Jonathan: Right, right. And I think that one of the things as I can get more people to think about in terms of a story, storyboarding, you know, you’ll get to that. So you better see what’s the problem we’re trying to solve? You know, who’s the audience? What action are we trying to take? And then what type of mechanism? Are we doing PowerPoint or PowerBI then, understanding your tone, stuff like that. Are we showing success or failure? Once you have all these questions answered then you can say, “Ok, how can I display that properly?” And then you can start getting to those amazing visualizations that Tufte and all those people show then you can start showing all those things. But once you have all that stuff answered upfront then you can solve. I think one of the problems that we try to jump to that is we say, “Oh ok, I think I can show that in this amazing histogram.”

Carlos: Yeah, exactly. I’ve got some data now let me just throw it in this visualization tool.

Jonathan: Right, and try to make it look pretty, and then it ends up not being what we wanted. It may not be exactly what we want and we try to tweak it. It can still in that being wrong. One of the examples that I use is you have somebody panting for gold out in California. And over the course of a week they have 10,000 tries and over the 10,000 tries let’s they say they find 147 pieces of gold. And so now somebody says, “I want a report of that”. And you say, “Ok, well what do you want in a report? Do you want me to show all 10,000 tries?” You may not want to see all 10,000 tries. You may only want to see the 141 successes. But you may be somebody who’s developing the tools that did the try so yeah you do want to see all 10,000 tries. So which is the right answer? And a lot of times too when somebody gives us a report requirement it may be three reports that we may have to develop, two reports. But we don’t know that until we actually go back and actually interviewing this and figure out who the audience is and stuff like that because at the end of the day it’s not just show us the data.

Carlos: Right, and we’ve talked about this before on the show and kind of maybe the new wave or the new technologist another differentiator is that I think a lot of times like you say,  we’re technologists I have the data I’m going to toss it into the report. I’ve done my job kind of a thing. The new wave or I think what executives are looking for. We’ve had CEOs on this show talking about what they’re looking for in data people. And they want people that understand the data. They want people who understand the business, right? Why are we collecting data this certain way? And I think this could then be a major differentiator to be able to go in and say, “Well you know what I am the guy who develop, use your Gold Rush example. I am the guy who, I understand how to develop this gold developing software, process whatever therefore I can give you better insights into that 10,000, or know how to display that kind of a thing.”

Jonathan: Right, and then two, know who your audience is for too because if you just don’t report for another person who’s going to be out there panting for gold that’s a whole different report than your CEO. Understanding all those things upfront before you ever open up the tool will help ensure that your visualization becomes useful because at the end of the day that’s the whole point to it.

Carlos: Right, that’s not the only problem.

Steve: So one of the things that I see working with clients specifically around reports is that oftentimes you’ll get a request that comes in from one of the customer of the client or an internal customer of the client that says, “Take an existing report that’s like a simple grid type report, and add to it or modify it or do these things to it.” And oftentimes what we run into there is that people know what they’ve seen and they know how to ask for alterations to what they’ve seen but they don’t necessarily know to ask for what they’re really looking for. Do you run into that similar?

Jonathan: I do and I have a couple of tricks that I will share. One of the things I do is like you said, you see that a lot. We have Report A, we want to extend Report A this and this. Once I have all these can I make a one sentence summary of it. Kind of make a three minute story of that. If I take what I have there and I give it back to them and they say, “No, no, wait. That’s not really what I want.” That’s what you will get when you start engaging the user by saying, putting stuff until a story is formed because we understand stories because we like stories. So once you start explaining that to them, they say, “Oh, wait, wait, that’s not it.”

Jonathan: One of the things that we always run across is that we think we know something and somebody may have, not that we don’t have, so it will be I think they know that. And it happens with all of us. We say, “Do such and such.” And we think that everybody may know all of the answers and they don’t. Putting it into a story form and presenting it back then we say, “Ok is this, and this…” And they’ll say, “Oh, wait I forgot to tell you such and such.” So you’ll prompting them to give you more information as well before you ever modify that chart.

Carlos: You’re now assuming that they know something that they may not know, or you know you’re working on different assumptions.

Jonathan: Right, because at the end of the day regardless of what kind of visualization you’re doing. What, you’re doing an infograph, simple chart, histogram doesn’t matter. You should be able to summarize that in words real quickly to whoever is presenting it to you and they just say, “Yeah, that’s exactly what I want.” Whether it’s the grain, everything should be summarized quickly to your user back and say, “Ok, yeah, that’s exactly what I want.” And at the end you could open your tool.

Steve: Sounds like some great advice there. You mentioned psychology earlier, can you tell us a little bit about how psychology plays into what you do with reporting and visualizations?

Jonathan: Yes, definitely. One of the major aspects of psychology is just color, right? What colors you’re using? And this goes back to my storyboard and you see everything comes back to storyboard, right because that’s like. Actually I have everything done before I ever open a tool. Even the colors picked out should be done before I ever picked a tool. So you go back to the tone of your visualization. Are you showing, you know, did you company lost $10 million last quarter? Do you really want to show that with happy colors and stuff like that? Maybe, right, you know. Right, maybe? But you probably don’t want the whole report to be red either. The color connotations of red, you know, how red makes us feel? How certain things we have innate biases to it. So like the color red, the color red has a connotation of power, energy, war, strength and stuff like that. So we use that in that aspect and that’s great. But it can also be violent, and brutal, overbearing, anger. It’s all about understanding the color that we use when we’re using it. And you see a lot of stuff, the visualization, you see the psychology of it and things like company logos, right? Like there’s a reason why majority of company logos have red, blue or white in it, if you’ve ever thought about that? Majority of company logos have red, blue and white because they wanted to show, you know, the red they want to show power, passion, strength, desire. The companies that have white want to show purity, and innocence, and simplicity. And the blue, like Microsoft obviously, right. They want to show confidence, and intelligence, and trust, stuff like this. So they’re already doing these things, you know, showing you these type of things. Even in movies there are certain actors, directors that use color to show feeling because we automatically have these innate feelings of a color when we see them so not going too far from the subject of visualization. We want to go back and make sure that we’re showing, that we’re using the right color. We are not using green because we like the color green. We’re suing green because we want to show nature, or down to earth, growth, vitality. But we got to be careful with it too because it can also be devious, and selfish and greedy as well. So want to make sure we’re cautious on how we use colors in the certain times. You can take a perfect visualization that shows exactly what the user wants and change the colors, and it changes the whole meaning of the report.

Steve: It makes me think I work for a company years ago where one of the marketing directors had basically a rule that nowhere on the company website would green ever be used. I never understood why but thinking about some of the things you’re talking about here maybe they have some kind of psychology behind not using green, and they fail to share that with the development team.

Jonathan: Well, what was the industry they are in?

Steve: It was in sort of self service websites like back in around 2000-2005 sort of early on.

Jonathan: Yeah, because I was trying to think of what kind of negative meaning has green would have or something like that, and they have them in something else that they were thinking about. Yeah, there’s all kind of reasons why people want to do that and then you get the stuff that’s. One of the funny things is like the color orange. Orange is historically a color that you either love or hate. But kids are drawn to it like that’s why Nickolodean’s Logo is orange, right. it signifies like happiness and stuff like that so you get things like Amazon’s logo and stuff to.

Carlos: Well, so it’s funny you mentioned that because people would ask me growing up like what my favorite color was, and I could never really pick one. But I always enjoyed like orange candies and flavor so I chose orange.

Jonathan: Because kids love orange, and this one of those things people don’t think about it until you’re like, “Oh yeah, you know what that’s true.”

Carlos: I think there’s one other area that I want to jump into but I guess I want to maybe stop and say. Well, look all of these sounds well and good and I think yes it make sense. Like we want to present the best data because at the end of the day we want to make sure that our user are happy, that people are understanding the information well. But all of a sudden this sounds like a lot of work. An additional task that I have to that now take on, right, in addition to this. Where’s my easy button, right?

Jonathan: Alright, so here’s your easy button for you.

Carlos: Ok, here we go.

Jonathan: Preventing rework, number one, because at the end of the day if your visualization is wrong you can do it again then you just wasted everything, right? Don’t think of it as, you know, I have to this extra work. Think of it as you’re doing it to prevent extra work. If you develop a framework, you develop your own storyboard and you get your stuff there. You develop your go to 5 questions. Because for me, like I have probably, maybe 8-10 charts and graphs that I use all the time. I already have those in my head. I can throw stuff into them and just go with them. I don’t have to think about those because I already have them. And if anybody who has done reporting sure you guys both have your favorite charts and graphs that you like. So you already have those, you don’t really have to develop them because you already have them and know them. Now if you develop your own…

Carlos: I don’t say, you know, showing the lazy man that I am, right? I mean, 99% of the time I’m taking the default, the default theme, whether that’s in Word, right. Whether that’s in like, you know. So whatever the chart is, I mean, like Excel might give me a couple of options but I was just like, “Ok, I want to take that one and then if they don’t like it then I’ll change it, right?” I admit that, and I think maybe to your point that idea of developing a process around how you’re going to create them and then coming up with some of your favorites. Once you’ve kind of gotten couple of winds there you can go back and make that process a little bit faster.

Jonathan: And you know what thinking of that too, what I’ll do, and I’m committing this now right on air so I have to do it. I’ll write a blog post of like let’s say 5 questions. A quick process that a professional can go through and may take them 20 minutes to gather these answers that will help frame a quick storyboard to build into their platform. It would be basic that you can extend it yourself but I’ll create a couple of questions that you can have and take in that way to help you get to that point so that you can eliminate a lot of the rework that we typically have. I’ll do that and put it in my blog.

Steve: Yeah, that would be great. If you give us the link to that we can include that on the show notes which will be at sqldatapartners.com/visualization.

Jonathan: Ok, yeah, definitely.

Steve: Ok, so now that we’ve talked about colors. What about people who are color blind? How do you handle that?

Jonathan: Ok, that’s actually a big big thing. And it’s actually dear to me too because I grow up with people in my family who suffer from Color Value Deficiency. And one of the things, I even remember as a kid and things that you would never think about unless you actually live with them or actually experience them in day to day life. Like one of the things are cakes, birthday cakes. I remember a birthday cake. We were doing a birthday cake and we have to actually stop and think about the color difference between the cake and letters on it because the person we were creating the cake for wouldn’t be able to see the difference. I was probably like 8 or 9 years old and I was like, “Wow, that’s…” So at an early age it was impactful for me to be able to see that.

Carlos: What are the stats on colorblindness?

Jonathan: I believe it’s something like 1 out of every 8 men are colorblind. 1 out of every 12 men, 8%, so yeah. 1 out of every 12 men and 1 out of every 200 women in general are colorblind. So if you take a SQL Saturday with 30 people in the room, there’s 3 people at least in that room that are colorblind. And then too, there’s not just one type of colorblindness. There are multiple spectrums. There is the Color Vision Deficiency, there’s different types of Protanopia. Those are the three major ones. But there are like Green Cone Deficiency. One of the things that we hear too when we think of colorblindness, we think that the person can’t see the color red but they lose the whole spectrum. So somebody has a green deficiency they lose the whole green spectrum. They could lose the whole red spectrum. And one of the things that I show in one of my talks is that, and I do this purposely because like as I said it’s dear to me, is I show people what other people see because until we actually see how other people see things we can’t really empathize with them. So I’ll take a picture and I’ll show it in the different types of spectrum. I’ll say, “Ok, this is how we see the world.” Then I’ll show, “This is how they see the world. This is how this person sees the world.” And they’re like, “Oh my God I didn’t think about that.” There’s actually some cool things that we can do though. Outside off the bat we can immediately, we can use shading and gradients which are tools that are already available to us to be able to show differences. We can avoid using reds and greens together when possible. But there are other tools too, like one is the ColorLab. And I will give you guys the links so that you can put it with the notes, the ColorLab. The ColorLab is actually pretty cool because you can go through and you take your company’s color palette because that’s a lot of thing too. A lot of companies have color palettes that they have to put their reports in these palettes.

Carlos: Right, they have to meet the criteria, sure.

Jonathan: Right, so they probably better use shading and gradients. But you can put the colors in here and you can go through and change the color spectrum, the CVD spectrum and see how people see these colors truly. And there’s another site, color-blindness.com, it’s called COBLIS, it’s the Colorblind Simulator. And so what I do is when get done with my visualization whether it’s my Reporting Services, PowerBI, Tableau doesn’t matter. I’ll screenshot it and I’ll upload it there and I’ll look through their simulator to make sure that I can still see what I want to show. Does my report, does my visualization still show the story that I want to tell so, am I loosing tone because I lose the color? And if I am then I can go back and change it. Those are quick things that we can do to make sure that we include people from the CVD spectrum because one of the things is that as time goes on we have more and more data and everything is becoming “datafied”. You’re going to have more and more people depending on data. And the one thing that you don’t want to do, right now it’s not written into the ADA, the American Disability Act, but I can see it being used later on because somebody say’s they can’t do their job because you don’t show them a CVD friendly report and they can’t do their job. And that’s actually something that is simple as changing the colors. You could be potentially liable in the future for stuff like that. I see those types of cases coming because I see lawyers chasing stuff like that. That’s a whole other spectrum to chase and make money. To hedge is bad. What you don’t want to do is to be the person who developed that report that got the company sued for $10 million. Right, I mean.

Carlos: No, that’s wasn’t me, that was the last guy, right?

Jonathan: You know, that’s stretching it, that’s an extreme outlier but the possibility exists. So just to prevent it let’s include them anyway. Because then too you will never know who you’re affecting. If you can make somebody’s day by including them without actively going out and searching them that makes somebody feels good. At the end of the day you can help somebody out. That’s what this is all about anyway, right? So 5 more minutes of your time to help include somebody who has CVD, that’s cool.

Steve: Who knows that person you’re helping out by making it look good for the colorblind is maybe the decision maker on whether you come back for that next project as a consultant.

Jonathan: Definitely, yeah.

Steve: That could make a big difference to your future as well as or my future.

Jonathan: Right.

Carlos: So what are some of the tools doing to help us with some of that, right? Again, I’m a lazy knuckle dragging Neanderthal. Alright, I want to kind of out of the box to be that way. What are the tools doing to help me there?

Jonathan: Well, I’m interested to see how PowerBI is going to directly address that. One of the things that you can do right now and like Excel and PowerPoint, and PowerBI as well, you can change gradients and stuff like that. I mean, right now, you got to be actively thinking about it. I could see in the future having a switch that if you have CVD, you could hit the switch and it would change for you. Actually that would probably make a lot of sense to do something like that because that’s your easy button, right? I don’t know if they’re going to do that but I could see them doing that because that’s simple. They could say, for this color, this RGB color we could show this color because they have the budget to be able to delve into that and figure out what works for who and what works for not. I could see them doing something like that in the future.

Carlos: Right. Yeah, just translating that so they do have the different colors like you mentioned the reds and greens where it just translates to gradients or shades of blue or whatever.    

Jonathan: Right, because the nature of the PowerBIs and the Tableaus stuff like that and then with the HTML 5 enable you to do stuff on a fly because it will then be just a quick button that will let you change the styling or shade something like that change the colors on a fly. Obviously a Reporting Service report is static. It would be able to do that. But the new HTML 5 reports will be able to do that. You know what I mean? I could see that coming in the future. We don’t know yet but they’re actively working on it so we know that they know what’s the problem and we will see something soon.

Carlos: Right, now will you talk a little bit about, again, kind of data everywhere, right. And you know SQL Server 2016, the integration with PowerBI and Mobile Reports. We may already feel that there are enough reports out there but it seems like that’s only going to expand, right? Will you talk a little bit about, you mentioned the dangers of datafication. I mean, you feel like the role of reporting and the need to go through this is just expanding. We’re now just starting to kind of get into some of this?

Jonathan: Oh yeah, this is the forefront of it. Like we’re datafying everything. We datafy everything from our footsteps, to our calories, to our heart rate, to traffic on the road, we datafy everything because we like to be able to see and analyze things. We want a quick answer. We want to be able to say, “Oh, I get this.” And it’s good though. It lets us make changes and decisions we can plan. They’re using it to be able to track diseases. They can track the spread of Zika virus using data. That’s amazing stuff you can do but there’s dangers in that though. There are extreme dangers in that. We need to be careful especially with us being the person that’s, the people that are actually showing the data because it falls back on us. I’ll give you two quick examples too. One is a mechanical issue and the other one is a human issue. And so at the end of the day as data professionals we love the term garbage in and garbage out. We live by that. So thinking along that I’m actually writing a blog post about it now so the story will be in there as well. I had a client that was a retail client and they want to track store attendance with a mechanical piece. And as every time someone came in it will track and count so they would know how many people came in and out of the store. So they would do everything, they would base stuff on store bonuses and stuff like that based on that number there. And they would based even their finances is bases on that attendance in the store because they could do store attendance divided by sales and all that kind of stuff like that. So as I was working on one of the analytical project there I was interviewing one of the business people, well I was interviewing the business people in general, and I begin to hear them call it the random number generator. And I was like, what? But it was known that the numbers we’re wrong and it was a joke. I’m like, wait a minute. Your basing your business on something you know that’s wrong. They like, “Yeah, we know it’s wrong.” I’m like, that’s not a problem? Everything they were doing in the future was based on these incorrect assumptions so the analytics was right, the visualization was right but the source was wrong. So you have garbage on the end because your source as wrong so that’s one of the dangers as having an incorrect source, having low quality input. The other one is the human factor and this is little touchy because how it transpired out but we’ll use the election. One of the things that we saw up until the point of the election, a week before the election, most major poll showed Hillary Clinton winning between 3-6% points. You know, ok that was great. And she had based the fact that, she didn’t have a concession speech because the numbers says she could not loose. Obviously we know how that went out. So the first thing that they went back to, they were like, “The polls are wrong, the polls are garbage.” So that reflects back on people like us writing the report so that will be the analog of we wrote the report wrong. No, we didn’t write the report wrong, the data came in wrong. What end up happening was it caused a social desirability bias. People says stuff what they think you want to hear not what they’re really are going to do. So people were saying that they were going to vote for her when obviously they didn’t. But like I said we won’t go into that, that’s also psychology as well. I told you I’m a psychology junkie. But see even with that, people knew we collected the data, they did the full random sampling. You have faces like the Gallup poll and USA Today poll so they’re doing all the stuff that’s correct. They’re doing the proper distribution of people, the proper distribution of the population to be able to get exactly what’s needed to be able to give accurate results and stuff like that. They go through, do all these and have the proper margin of error, you know, everything, right? And they showed, “Ok, 3-6% she should win. Ok, great let’s work on that.” And then that doesn’t end up happening because the source was wrong. One of the things that I want to caution is as data professionals we need to make sure that we’re comfortable in our sources. Is our source correct? Was there a bias in collecting it? Were there errors in the machine that collected it? Was there possibility of errors coming in with the data corrupted coming in, all kind of stuff like that because at the end of the day…

Carlos: Are we looking for the number that will help us prove our point rather than looking at the numbers a little more objectively to help us answer the question.

Jonathan: Yeah, I think so and that’s a great way to put it. Are we, in essence that’s another bias. Are we using the numbers to prove our bias or we’re using the numbers to disprove our bias and to show the proper thing because that’s the end. Like you said, they were using the numbers to prove their bias because they wanted her to win so the numbers proved that so that works let’s go with it. What nobody ever said was, are people truthfully telling the story that’s actually happening. Are they lying to us? Are they truthfully going to vote for her.

Carlos: Right. Man, well some interesting stuff and I always see, I think. This is not going away, right? The reporting options are not decreasing that are available to us.

Jonathan: Not at all.

Carlos: And so I think it will be kind of interesting to see how some of these plays out. I’m interested to start playing with even some of the colorblindness components. Because when I heard those numbers I thought, “Wow that is much higher than I would have guessed.”

Jonathan: One of the things too with the influx of data and how it’s going to become. I could see these principles being used for everybody from the Administrative Assistant to the Executive Director and CEO and stuff like that because everybody at some point is going to become natural to be creating reports. We need to make sure are we comfortable with the source because it’s not just going to be us creating them in the future. As more and more self service tools come out, you know, the iPhone 15 is going to have a way to show reports of your tweets and stuff in real time. You know, Facebook is going to be like, “Hey, show my data.” I mean, I can see that. That’s coming because it is just the next natural evolution of it because as more and more people understand how valuable data is the more and more people are going to show their data. More and more people are going to show what they’re doing with their data. I mean, think about it. Even your FitBit, you know the FitBit, get your own reports so that you can share with your friends. You know, “Look, I had 10,000 steps today. I’m averaging 9.2 hundred steps a week.” You know, stuff like that. So even those small stuff, even now we’re beginning to see people wanting to show their own data. So it’s just going to get more and more in the future as we get further along.

Carlos: So that’s an interesting point. And I think ultimately as data professional we can continue to provide value there if we’re willing to educate people on how to do that and do the best practices rather than I am the creator of the reports, here’s some guidance on how to do that.

Jonathan: And I think that’s the big thing at the end of the day is willing to see control for the greater good. Because at the end of the day you’re more valuable when you provide true benefit to the enterprise than just you know, “I created this”. If you can show more people to do it better than just having, because you’re a bottleneck no matter how good you and me, we are bottlenecks. But if ten of us can do it then we’re better. That’s how I see that.

Carlos: The value there again is an understanding the data and kind of where it comes from and then the presentation that’s like the final piece. You can push that down a little bit and just have that be taking off your plate as the tools become easier and what not. Yeah, good stuff.

Steve: I think what is interesting on that too is that oftentimes the visualizations or the reports that are shown will drive behavior. And when you’re asking someone to build the report that’s going to drive behavior oftentimes they don’t know the complete understanding of that behavior. And I think that goes back to what Jonathan said in the beginning around the story, building the story and storyboard. But if that is pushed out to everyone who’s able to get to the data and build their own visualizations then people really have to build their own stories in that case so that they’re showing the right data. Otherwise, you’re going to end up with data that’s driving business decisions that’s all over the spectrum. One person’s report may contradict another’s based off of their biases or approach to creating that.

Jonathan: Correct. And another thing too with reports this is something that I learned a long time ago too is that a report does not have to answer your question. Sometimes a report makes more questions which eventually lead you to the right answer. That’s something that lot of us do is that we want to report to solve a problem. Of course it doesn’t always have to solve a problem. Make them lead to more questions and that’s progress. That’s what I was thinking about.

Steve: Yeah, that’s one of the things I love with reports is when somebody looks at it and say, “Well, why is that number that way?” And you have to do the research and figure out, “Oh, well it’s because sales are off in that region” or whatever it may be. And then you’re able to fix a business problem because of those questions that came up.

Jonathan: As a consultant when I’m talking to clients that’s what I tell them too. Don’t be afraid to have your results posts more questions because that’s good. You’re getting further into your data. You’re learning more about yourself. Everything doesn’t have to have a quick answer. You know, that’s our society, we want quick answers. But everything doesn’t have a quick answer.

Carlos: Well, awesome Jonathan. Thank you for the discussion.

Jonathan: Yes, thank you. I enjoyed it. I love talking about stuff like this. You could can geek me out for hours for psychology and stuff.

Steve: This has been a lot of fun.

Carlos: Shall we go ahead and do SQL Family?

Jonathan: Sure.

Steve: So Jonathan how did you first get started with SQL Server?

Jonathan: I have a funny story with that. I was working on a help desk and I was trying to figure out what I want, I know I wanted a career in IT, because I wrote my first program when I was 6 years old. I was one of those weird kids. I had that Radio Shack TRS-80. I’m taking it back, right, showing my age. So as 6 years old I wrote my first program so I had been doing IT stuff all through school and stuff like that then got a job at a help desk and I knew that wasn’t really what I wanted to do. So I was doing research and research and end up finding, you know I was like, “I think this database thing is kind of cool.” So I went and sought out the DBA at the company and I work 2nd shift. He worked the regular 8 to 5. So I would come in everyday at 8 o’clock and sit with him until I had the work at 2 to learn SQL Server from him. And he was willing to actually, his name was Carlos too which is funny.

Carlos: Great name.

Jonathan: Yeah. And he was willing to sit with me and teach me SQL Server, and that was SQL Server 6, 6.5, that’s taking back too. They saw my drive and determination with that and the created me a Junior DBA role. And that was my first job as a true SQL Server professional, I was a Junior DBA at that company, and that’s how I got started. That was my first exposure to SQL Server and ever since then I just love SQL Server. I’ve been doing it for, if I tell you how many years ago you’ll know how old I am, so 17 years later this where we are today.

Steve: That’s impressive, going out of your way to sort of work the extra hours so just you can learn something new. If more people did that I think we’ll be in a whole different world.

Carlos: That’s right. Not the typical path, right, kind of falling into your lap, you know with me. It’s more of a, hey that’s pretty cool and let me go check that out. Now, in all that time with SQL Server, if there is one thing that you could change about SQL Server what would it be?

Jonathan: The one thing I will change is that they’re actually changing it. It was, you know, the staticness between versions and fixing things. We went to that whole dead zone, the desert between 2008 R2 and 2012, and that was like, “Oh my god”. You know, trying to figure out, and there were some great things professionals found. Ways around, solutions around problems but that was one of the biggest things between there and there. And then 2012 to 2014, ok that’s cool but to see them how faster change stuff like VNX is in CTP1 now. Is that what it is? They are at a rate where they could possibly even release VNX this year. That’s crazy and that was actually what I’m thinking was, what’s to be able to make changes faster on a fly is the community. And you know, users will say, “Hey, do this. Can do that?” You know, we’re getting that base for PowerBI seems like every week. It seems like every time I open up Management Studio there is another update for that. I mean, they’re doing the things that I would like to change and it’s changed faster. So that’s really cool to see as exciting. I can’t wait to see the future. I mean, imagine SQL Server, well the changes they’re making in just 3 years, 2-3 years it’s going to be amazing what’s coming.

Steve: Alright, so what is the best piece of career advice that you’ve received? So for me the best piece of career advice is based on, for me I had struggled personally in just being comfortable on who I am. And I remember one of my managers, his name was Keith. And me and Keith are still really good friends. I talk to him more of the time, I still do. You know, Keith told me, “Just embrace who you are. Be who you are. Don’t try to be somebody else.” And what’s funny is that’s actually the beginning of my logo. You know, you guys see my logo it’s my hair. That’s me literally embracing myself. And so that was the best piece of advice that I got because once I truly embrace who I am, my strength and my weaknesses. Still trying to fit into a box that I thought somebody else wanted me to be that’s when I begin to flourish. That’s when I begin to be truly happy. I’m truly happy because I’m being who I am. I love to share, I love to talk, I love to teach, and just embracing who I am. The best piece of advice I could give back to somebody is to tell them, “Doesn’t matter who you are. Whether you’re short, tall, you are good at this and good at that. Be who you are, embrace who you are.” And that will be your gift. That will be what people will gravitate towards you.

Steve: Sounds like great advice.

Carlos: You know, that’s right, that whole idea about networking kind of finding your own people. I think all too often when we come out of college and just like, “Oh, I need a job, right, so I’ll go anywhere. If we can identify what it is we want to do or place we want to go after. I mean, even this idea of colorblindness and making reports better to help that segment that kind of speaks to the work that you want to do and just a matter of finding where that’s important to other people as well.

Jonathan: Right, definitely. I think that’s even, actually even people taking that project and run with it. Someone selling a service to a business saying, “Hey, you know, we can convert your reports to.” Somebody say it’s like that you can sell that service.

Carlos: Our last question for you today Jonathan. If you could have one superhero power, what would it be and why do you want it?

Jonathan: So my favorite superhero doesn’t even have any superpowers. You can guess who it is?

Carlos: Batman?

Jonathan: It’s Batman, yeah. And what I always love about him is the detective comics because he uses his mind and that was a superpower. He just outsmarted people. Obviously he was physically fit, right. I mean, the ability to get to the in-depth of stuff. I love that investigative and that’s kind of like what I want to do anyway. Get to the problem, get to the root problem. Figure out what’s really going on. If anything I would want to be just Batman that’s my. I don’t need super speed and super strength. Nah, I just like the super mind and call it a day.

Steve: You know the other cool thing with Batman is he had pretty much unlimited budget, and he had all the toys, and tools, and gadgets that he wanted to do whatever he needed to.   

Jonathan: That is true.

Carlos: That will be a lot of fun.

Jonathan: Yes, I guess you could say that’s also super power too, right, the unlimited money.

Steve: Yup, right.

Carlos: Having a kin business sense or at least having a dad maybe who had one.

Jonathan: Right, right. That would definitely help.

Carlos: Well, Jonathan, thanks so much for being with us today. We do appreciate it.

Jonathan: Yeah, once again I thank both of you guys. This has been an honor to be able to do this. I like to thank the community for even, having you guys available to do this so I look forward to what’s coming in the future. Look forward to seeing you guys again at the SQL Saturday being able to panel and stuff. And I thank you for the opportunity and the time. You guys have a great day!

Carlos: Yup.

Steve: Thanks Jonathan, it’s been great.

1 Comment

Leave a Reply

Back to top