Podcasts > Operations > Ep. 046 - The medium is the message: data visualisation for a digitally-enabled organisation
Download MP3
Ep. 046
The medium is the message: data visualisation for a digitally-enabled organisation
Kirsten Sosulski, Associate Professor of Information Systems, NYU
Monday, February 11, 2019

In this episode, we discuss the role of data visualisation, why we use it, and how it is changing based on new technologies. 

Is data visualisation a leadership skill? How do we create a data practice within an organisation? How should we think through the role of data visualisation in your organisation?

Kristen is an Associate Professor of Information Systems at New York University’s Stern School of Business. She teaches MBA, undergraduate, executive, and online courses in data visualization, computer programming, and the role of information technology in business and society. She is also the Director of the Learning Science Lab for the NYU Stern where she leads team to design immersive learning environments for professional business school education. 

Kristen book: Data Visualisation Made Simple https://www.amazon.com/Data-Visualization-Simple-Kristen-Sosulski/dp/1138503916

Twitter: @sosulski

Linkedin: https://www.linkedin.com/in/sosulski/ 

Email: ks123@nyu.edu



Erik: Welcome to the Industrial IoT Spotlight, your number one spot for insight from industrial IoT thought leaders who are transforming businesses today with your host, Erik Walenza.

Welcome back to the IoT spotlight podcast. This is your host, Erik Walenza, CEO of IoT ONE. And I'm joined today by Dr. Kristen Sosulski. Kristen is a data visualization professor at NYU Stern School of Business, as well as an author and speaker on the topic. And today, the topic at hand is going to be how management can make use of data visualization technologies to derive insight and make decisions based on the absolute surge of data that IoT systems are now providing them. And of course, the systems are only providing them this data in theory in many cases. So the data exists somewhere in a database, but without proper data visualization solutions, management is often unable to make decisions or able to make decisions only with a big time lag, so unable to make real time decisions based on the data.

I hope you enjoy the conversation. And I look forward to your feedback. Kristen, thank you so much for taking the time to speak with us today about data visualization.

Kristen: Thank you, Erik. I'm so excited to be here.

Erik: So Kristen, I've given a quick intro to you already in the intro. But maybe before we launch into a conversation, you can just take a quick moment to introduce your recently published book “Data Visualization Made Simple” as kind of a foundation for our conversation here. So here we'll be talking about the role of data visualization, how visualization is changing through the adoption of new technologies, and the impact that that has on our relationship to data, data visualization across times across different audiences and formats, and the importance of establishing a data practice in your organization. Maybe ground us in your background. What was the inspiration for publishing this book? Why is this important at this point in time?

Erik: Absolutely. So data visualization is essential skill required in today's information-rich world. And from anyone from like interns, MBAs, managers, it is a leadership skill. And through my teaching and professor at New York University's Stern School of Business, my students have actually challenged me to create the business case for why visualization needs to be part of a firm's practice. And I'm so passionate about this stance on being able to communicate insights from data clearly across the organization from data folks to non-data folks. And it's really as simple as that.

And so, my motivation comes from my really passion for education and to democratize the access to information within an organization to improve everyone's decision making, and ultimately think about the customer experience.

Erik: One of the most pervasive topics that comes up is capturing data, breaking down data silos. All too often, the topic then of actually making this data accessible to the human beings that need to make decisions based on it is not a significant part of that conversation. So it's okay, are we capturing the data? Are we able to analyze the data? But then when you talk to the operators of the people that are supposed to be making business decisions or operational decisions based on this data, they often are caught in a last mile problem where they then don't have really sufficient access to this data at the time that they need it.

Kristen: And that's why it's so important to not only have a visualization practice, but establish a data practice and build that into the culture of your business.

Erik: So let's start with this foundation of the role of data visualization. When you're communicating to your management students, how do you think through the different roles of data visualization in a modern organization?

Kristen: Like there's different roles for the use of data graphics in an organization. So it can be anywhere from exploration. So when you're trying to build a data understanding or understanding of your data, a lot of times we use visualization to be able to spot like correlations and spot trends and have those preliminary insights that help us build our questions that drive or analysis. And so, data exploration is a much different activity as you know than data modeling.

And data modeling, there's many visualizations that help us like different correlation matrices and such that allow us to better understand the accuracy of our models, and, and so on. And so those data graphics are used for a specific purpose. And then we go to like, how do we use data graphics for decision-making? There, you're communicating to a broader audience. So data analysis to managers, and how managers can easily interpret data graphics to make decisions and have like the most important information today.

And so there's different parameters and different inputs that are required for that and also, different visual cues that we can use for that product, have much more simpler displays, ones that really just show the key information that's needed to help those in their roles within an organization.

And then there's the idea of presentation. So now you're not like using inputs to understand what's happening now in organization, but you might be trying to predict what might happen tomorrow. And so in that case, you're really building the case for why your predictive models are showcasing that there could be new markets tomorrow, and how we have to act now, for example. And so when you're building pitches, using visualization to show the evidence of the data, and all that strong analysis that you've done is critical. But people's eyes are going to glaze over when you start talking stats immediately. And so what you want to do is use the visual as really evidence, like pretend that you're really seeing a table of numbers but you're really seeing a data graphic.

Erik: So if we consider maybe data visualization used for exploration, and then used for making very specific decisions, and so we have these different purposes, how would you then advise a company in terms of the structure of the visualization, or who might be working with that to accomplish these two different purposes?

Kristen: Well, one is that we need folks to have an understanding of the data and what that data represents in the real world for the business that requires like an understanding of all the different columns and rows within your data structures. And to be able to see if the questions that our pose say by management are actually something that we can answer with the data. And so that's a different type of skill than one in which we are presenting the results of an analysis to a broader audience.

And that requires really well, let's think about how as humans we retain information, let's think about humans how we can perceive these various data graphics and their encodings. For instance, if we're talking about like a lay audience, we may not want to show data graphics, like histograms, or box plots, or parallel coordinates, or things that are might be more statistical in nature, or more complex. We want to show data graphics that are easily interpretable. It's not about trickery here or complexity. It's about showing the key insight, and using, again, that to provide evidence. So that's really the difference.

Erik: So to an extent we're telling stories, but we're telling different audiences for different purposes, and we need to then couch them in different formats. We also have recently a surge in new technology that's allowing us to visualize data on the one hand in different ways, visualize different types of data or mixed datasets that previously we didn't have accessible, and also view them on very different formats. Talk to us a little bit about the change that you've seen in terms of data visualization, technology, and how we then as organizations, and as people interact with that in the past, maybe 5-10 years? And then do you see significant change on the horizon in terms of how we're interacting with data?

Kristen: Absolutely. I mean, so there's a lot of questions there to unpack. But why don't we start about like some of the changes? In my book, I talk about like two major trends that are taking place, that are affecting the field of data visualization. Like one, which you mentioned earlier storytelling, this emphasis on data stories. And why? Because we understand that data is just on its own is pretty much meaningless. And so we have to transform that data into information and hopefully knowledge that someone can act on.

And doing that we can situate that in the context of a story, especially when we want to be persuasive or when we can pitch, or we're presenting something that might be futuristic like a prediction, rather than just reporting on the past. So I always think the most boring data presentations are when I just learned about the past and there's no suggestion about what might happen in the future, it's just simply a report. But those ones that are particularly compelling are the ones that like inform you and inspire you to take action of some way. So, that's one thing that has definitely changed with the field of data visualization.

A second is the ability to create interactive data graphics. So now I don't always have to rely on say, the presenter to have a presenter-driven story. Now there's audience-driven storytelling. And you think about that. I can input some data into a data graphic, like my zip code, or like my gender or something about me and I can see the output of that on how that relates to you to my own personal narrative.

So if I want to look at how people spend their time, there's a lot of like interesting visualizations about how people spend their time conducted by the US Census. And so I might choose like female between a certain age range who was employed. And I might see how like all females within my age range like spend their time working versus exercising, versus watching movies. All those different things. So now that the information is made much more meaningful to me because I can explore it through the lens that I choose. And that lens is a little narrow because the designer of the data graphic picks those filters that you can use to explore. But that's creating experiences now.

And I do a lot of training, consulting and teaching around data visualization, believe me, there's plenty of work to go around. Because you're asking statisticians in one way to become interactive, graphic designers, if you really think about it, to think about the holistic experience, and to think about an audience in such a different way that we ever have before when presenting different data graphics. And so I think this is like a huge opportunity to almost have a two way conversation with our audience. It's a much more powerful way to actually learn from the data, and to encourage this this line of inquiry, anytime we present data graphic.

But it also requires that the audience is mobile. When they're in front of us, we're not going to allow them to have that interactive experience. Otherwise, why do we even have to be in the room? And so, there's a lot of different like considerations one has to make in terms of the format of a data graphic, like whether it's on your little watch, or it's designed for desktop, or something that's going to be printed as a report that I can read on subway? Or is it going to be something that I need to be present for a live presentation?

Erik: So one of the things that we try to do when we're running a workshop is we'll often try to have a quick 32nd, one minute survey of the participants because having that immediate input and being able to see on a screen, so if everybody's maybe using a WeChat app and voting through their mobile phone, but then they can then all of a sudden have a little bit of insight into the experience of everybody else in the room. And somehow that is because it's immediate, and it's related to the people that are directly in their vicinity. It's more impactful than if we just said, here across China are the average results from a survey that we did last month.

Or if we're looking at visit Alibaba or something, and they have a great visualization, where they can show in the past 24 hours, how much traffic has gone through their servers, and how much ecommerce has gone through their sites and being able to see that this actually occurred in the past 24 hours, this is kind of real time data, it's much more impactful again, than say in the past year, on average per day, we did this much volume. So somehow this this immediacy of data also has a very different impact than looking at averages that are then just brought down to a particular unit of time.

Kristen: The immediacy adds this notion of privilege. I'm seeing this special information right now that how many people could be seeing at this one moment. And I think it's adding a level of intimacy into an organization or phenomena that wasn't possible before. So we're actually getting closer to what's happening now, as opposed to like really reflecting on like trends that happened in the past. And I think this is like so powerful. And how we start using this information and expecting it in some regard is definitely going to change, I believe us as humans, just like our behaviors, and how we consume information, and how it affects the scale and pace of our everyday activities.

Erik: You mentioned earlier the ability also of data is not good or bad, true or false? Well, I mean, hopefully, it's somehow true. We have right now unfolding a great case study here, which is the midterm elections in the US. And now if you look on any online media in the US, there's going to be visualization in real time of results coming in. And over the past couple of months, we've had different organizations telling different stories with different data points, without getting overly political left or right here, but just the context of data in an environment where it is used to convey a message and win an argument to an extent.

How do you perceive this because this has been quite powerful, whether we're looking in the political domain or as a maybe an investor, an early stage startup selectively choosing data fitting them into a story in order to secure their next round of investment, you again have this really powerful and modestly manipulative you could say, use of data to convey a message?

Kristen: So, I have a couple of points. So my first point is, as data graphic designer, you are in complete control of what you show and what you don't show, what you present, how you present it. Do you show all the information, a partial story, and actually how you visualize the data specifically? It's really easy, even though it might be computer generated to lie with data and to manipulate that and to have the graphics look as though they are presenting reality when, in fact, they don't.

And anyone who's a novice at creating data graphics, and that has evolved in their practice and as more of an expert today knows that it's very easy to make mistakes with data. So if we know that we can accidentally do it, we can definitely do it on purpose. And we can definitely create data stories that serve our purpose. Data could be used to advertise, to persuade and to promote. And that's part of the power of it. We're providing evidence. Now we can think about the context in which that data was collected, the ethics that were used to collect that data, the privacy of the participants who contributed that data, and so on. So there's a lot of ways that in lenses we can look at it, especially when we're talking about using data for political purposes.

But the key here is like to point number one is that we often think that okay, we credit data graphic, and it's kind of out of our hands. But the point I want to make is that the designers are in complete control, and completely responsible for what show. And they should 100% know what that information means and what it means in the real world and try to figure out what the impacts going to be of that.

The second point I want to make is, in part of my book actually, there's a professor of political science at Columbia University, who has actually written a small case for my book. And it's called “Policy Mood and Partisan Control of the Presidency”. And he talks about, and he shows a visual that shows the different cycles that public opinion takes with respect to demand for liberal versus conservative policies.

For some of us, things might be looking up. As we get towards, more conservative policies, the momentum usually changes to more liberal ones. And we've seen that year over year and he shows that as evidence in my book through a data graphic. And so that should be reassuring to some, but also to kind of be able to reflect on what the things that are happening now, in terms of every single number, every single input that we're getting and reporting on, the status of the midterm elections and all these things that they will add up to a whole at some point. But at some point, I don't think we always have the whole information. And we might be looking at certain distinct data points. But we don't necessarily have the entire population represented in that.

And so being able to couch that in these different historical theories is pretty important. We can look back on things that have been visualized about the different policy mood of the nation and how that's changed from Obama onward.

Erik: You'd mentioned that the creator of the visual aid station is going in control of the story, they're in control of what they share. In the political context, also in the business context, we're then in a situation where we're constantly being presented with storylines backed by seemingly very viable data that is often telling very contradictory stories. What suggestions do you have to our listeners as consumers of this data?

Of course, we then have an a responsibility as voters or as business decision makers to also up our game or improve our ability to translate the data that we're receiving into rational decisions so that we're not just, you could say, only accepting the story that is being told to us but also combining different stories are providing our own input into that storyline to enrich it. I think we certainly have, in some contexts right now, a bit of an overload, where we're grappling with the challenge of how to process all of this data, and be able to then make rational decisions out of it. So maybe, from the perspective of the receiver, what advice would you have?

Kristen: So there's this whole field called media literacy. And it's been around for a really long time. And this really studies the impact of messaging through media on our culture, and how to decode what is being said, truth from fiction or a way to develop a critical eye for the information that is communicated to us and not always to take that at face value, but to evaluate that in terms of a larger context.

I feel like we all have to just have a set of research skills. I have a doctorate, I got research, and so like, what do you learn when you're a researcher? Like you know how to sample from a population, and you know what a good sample is, and you know what a bad sample is, you know what a representative sample is, and you know when that's not, and being able to understand these basic principles will be able us for it to develop this critical language that we so very much need in order to kind of decode all this information that we are consuming, and the integrity of it.

For anyone, anytime you're looking at data graphic, look at the source of the data graphic. If there's no source, you have no idea where that information came from. So that's like, step one. If there's no source like you shouldn't trust it. And then if you also see there's a source, is there a date? From what time frame is this from? And so somebody's showing you something from 1973 trying to talk about now, that's a point of critique and question. And so these are just some starting points.

Erik: And here, we have an interesting dynamic where on the one hand we have large datasets over time across regions that can be presented up to senior management for longer term decision making towards middle management for more ongoing decision making, and also to operators on the ground for really real time reaction to data. And here, we're often looking then at data that's coming from the same sensors or the same systems. But the data that people are interacting with is over different time periods, over different geographic regions. And it's certainly presented to them in very different ways.

I know ThingWorx as an example, let's say they are working with Caterpillar or somebody, so they want to be presenting this data up to Caterpillar’s executive management, they also want to be presenting it to the maintenance teams that are doing maintenance and responding to real time orders. How do you think in the context of this organization in terms of structuring the data and putting it in the right place at the right time in the right format for each particular audience out of, you could probably segment a large organization into really dozens of different distinct audiences that have different data requirements?

Kristen: Yeah, I think it's called Magic, and perfection. And so, I mean, it really depends upon like your organization. If you already have a culture where you have equitable data access and you also have your data isn't siloed and you're able to leverage the use of all these like nice BI or Business Intelligence tools, like you can do this really easily, right, you can set the access levels, and you can build different reports that makes sense for the users.

It's not just like build it and walk away, and hope people are going to make sense of the information. It's like it's about incorporating that into the practice of reporting. So anybody that's like on the ground using this information like day to day to the manager that they're reporting to the business unit VPs and C-suite executives on how they use this information, and why it's important, I think that's like the biggest component of this is figuring out what is the most important information, and how best to communicate that to help with decision making.

And then also, when reporting back up, what's the most important information that management needs to know? And circle back that this information is actually being used in the business strategy, for example. And I think having that feedback loop is really important and reinforce it. And you can jumpstart the culture by having influencers in each of the different business units, and having champions for establishing this data practice, and the ability to use these different data insights to make decisions become smarter about your business.

Erik: So you've mentioned earlier this concept of a data practice, and I guess organizations have IT departments, they have CIO functions, they have CTO functions that may overlap in some areas in terms of how technology is used internally. But none of these, I feel, at least, their primary responsibility is from the user perspective. A lot of the IT department is often functioning from the perspective of make sure that the infrastructure runs effectively, not so much make sure that the users are satisfied with the data that they're getting and the way it's visualized. What exactly then do you mean by a data practice? Or what precisely would this look like in an organization? Where would it fall in an organization? Who would be involved? How would it be structured?

Kristen: We know with like the rising popularity of like data science, business analytics, like the skills in analysis, statistics, and data biz are like in high demand. And given that building a strong analytics and visualization team can definitely help affirm transform their data into unique asset. But it doesn't mean that you need to have your own data science or analytics team to run this.

But it is a quite different task than your CTO. It's not just making sure the servers run. It's making sure that the models and the predictions that those models are outputting are accurate, and reliable and actionable, and that those results can then be communicated to the people that are going to be using them as inputs into their analysis and decision making.

So, by having a group that understands the data in which an organization is collecting, and those things that are absent from their data collection methods that are questions left unanswered because maybe they don't have access to the data, or no one thought about, like, let's try to collect that, or that data is siloed, and it's hard to merge it together, all those things are really real problems in going from being technologically savvy to being data savvy.

And so knowing more about our customers and our customer actions, and what they're doing is absolutely critical. Like even in higher education, like knowing like what my students are doing on their learning platforms is much more helpful than not knowing. But if I know that they're accessing a lesson and watching a video that I created, that's like a data point for me. And knowing who did and who didn't, that's another data point for me.

And so we're able to aggregate all these data points to have a picture of what's happening today in a much more nuanced way than ever before. It's not something our CTO is going to be able to tell us. Maybe our marketing team will be able to tell us like clicks and things like that, and conversions. But if we're looking for a particular like path to purchase or how customers are interacting with our platform, that's a much different set of set of questions. We might be running AB tests to capture that data, which is like really common in organizations like Google and Amazon, like part of their practice to have this line of inquiry, constantly testing new ideas running, ad A versus B, and analyzing the results of that. That's not something your CTO is doing. That's a combination of your data analytics team and your marketing team.

This is just like the tip of the iceberg, which is thinking about the possibilities, the types of questions and how you can grow your organization by learning more about your customers in the present, rather than just looking at past information, just because you're able to collect it. We can actually be active data collectors within organizations. And then your create creating a plan a plan of action, and determine who's going to have access to the insights, how they'll be used, looking for measurable outcomes, ranking and prioritizing answers to the questions to make better informed decisions.

And then you have your whole culture of use that you're trying to build. You want the access to data to be easy and manageable. So if it's still hard, no one's going to be looking at it. And that means actually thinking about who's going to be using it, thinking about your audience. Other times we think about our audience that’s like customers or clients. But if you think about your audience as your fellow colleagues, like how can we make the organization as a whole better, and that's a different type of mindset.

And then training is necessary in like how we user communicator results and use data as evidence and how we present with charts and graphs? It seems pretty simple, like tons of people consult on this practice, like presenting with charts and graphs. But everything by you put a chart on the screen, and how many people don't talk about the charts they have on the screen these days? Or they have five of them, they only speak to two of them. Or they're just meant to present like a pretty overview. But they're not spoken to. And the audience is left wondering, like, what did that mean? Why was I looking at that?

And so all these things is like really just taking your analysis and your understanding the data outside of your own shoes, and thinking about the folks that are going to be viewing it and interpreting it, and how to design best for them. And then measuring the impact of all this, and knowing that what you do today is going to look very different tomorrow. And I speak about that which is the subtitle of my book, which is called “Insights into Becoming Visual”, meaning that it's not just like you read a book on database, and now you're an expert. It's a practice. Just like anything, I'm a big tennis player: just because I played once doesn't mean that I'm an expert at it. It's a constant struggle.

So thinking about how this is the extra 20% that you put into all your data analytics work, anytime you're working with data, figuring out how to best summarize and synthesize it for others.

Erik: And also, I think, for yourself to an extent this concept of Kaizen of continuous improvement in an operational standpoint, this, I think, actually makes a lot of sense in terms of how we interact with data today. So, if the decision making is isolated to a digital office, or some particular group of people in the organization, okay, those people can develop a very deep level of expertise in terms of how to actually construct visualizations, analyze data, and so forth. But it's often the person on the ground closest to the activity that really understands what would be useful.

So if I had this data presented to me at this point of time in this way, I would be able to do my job better and make better decisions and so forth. And that information is often simply not visible to the experts, the people who are really have expertise in terms of aggregating and managing data. And so, somehow, we need to also get the organization at large up to a certain level where they are able to identify what data is potentially available, how that data could be visualized, and how it could be used and to be able to filter that back to advise on company operations in terms of what is actually made possible.

This is just, I think, a constant challenge where you have a particular almost you have a brain center somewhere in an organization that is great at generating ideas and managing data, but is fairly disconnected from the endpoints of the organization where the work is getting done and somehow we need to engage those people who they might be have a fairly low education base. They might be scattered in different regions speaking different languages across the world. Do you have any case studies or any perspective to share in terms of how do you bring data visualization out of headquarters and bring it down towards operations towards where work is getting done so people are able to be more engaged in terms of what data they receive, and how it's presented to them to improve their ability to do their work?

Kristen: And what you're essentially saying is like, hey, there's an opportunity out there, that if you're great at visualizing data, there's a huge opportunity for you on the job market right now. Because firms need this, they need people that are skilled. And it may not come from headquarters. It may be something that coming from the ground up. And that's one of the things that I teach my MBA students is that this is really a leadership skill. But if you're able to communicate what you've done, and persuade, and show how you've done it in a way that's digestible, and understandable, and the word I love, it's interpretable, I mean, that's a huge win.

And so when you're talking, Erik, about how we communicate through all levels of the organization establishes data practice, I mean, there is one going case study that is in my book which is about the gaming industry. And it's a story that that really speaks to, one, data being siloed. So you might have things happening in the marketing department, and you might have things happening in the game rooms of a casino and maybe those two groups aren't talking to one another. So that's one part of the case study.

The second part is you want to make a decision about okay, well, we have only have a certain amount of square footage in our floor and we're thinking about closing down the bingo room, and putting in more slot machines. Why? Well, I mean, it's a simple like numbers game, slots bring in more money than bingo. And the case goes through, like, running what's called like a cluster analysis and being able to identify the different segments of our customers, and being able to help like the floor manager specifically see and understand, like, who's actually playing bingo? What else are they playing? Who's playing slots and what other games are they playing?

So if we close the bingo room, what other business are we losing? Are we also losing business in the slots? And so what's the total loss rather than just thinking about only the opportunity, and I think that's something where having a more holistic understanding of our customers is super important. Again, this is at the gaming for level.

If you've been to a casino or even Target, you have things like these reward cards. So you put a few quarters in the slot machine, and you also put your card in. I know that you're in Shanghai, so let me give you like a Macau example, like going through the wind [inaudible 38:17]. And it's recording that information. What else did this type of user play? Being able to kind of see that, and see that on a daily basis and see how it changes from weekend to weekday, and if you're managing staff how to staff up, how to be more responsive, how to know who's there? Do we have what they call like a big whale, like somebody who spends a lot of money? Like do we know that they're there? And so there's lots of, at all levels, on being able to ensure that the folks who need the information are informed.

Erik: You'd mentioned also when we were chatting before the call this online certification. I'm thinking that going through an MBA program and so forth is something that only so many people can actually make that investment. But there's, of course, other lighter ways for people to get their foundation and be able to orient. Maybe you can share a little bit of input on this and help us understand who would this be useful for? Or what are the other mechanisms for somebody who is not yet ready to dive into an MBA and devote a lot of time to this but still kind of wants kick start in order to accelerate their ability to manage data?

Kristen: In my NYU Stern, we are offering a certificate called Visualizing Data; it begins in February of 2019 and ends at the end of March, 2019. So it's only eight weeks. It's completely online. And there actually are some live synchronous sessions that are optional. So you actually get to talk and interact with me. It's all based on my book. But it's about what I'm calling the practice of becoming visual.

But the course really is what propels you into actually developing the skills, like how to build a data map, and what that looks like, and how to make sure that data map is representing our data in a normalized way, and accurately, and all that good stuff. And so we really get down to using the software, but always keeping in mind who our audience is. And so actually, the assignments are all about creating visualizations that represent the past, the present and the future.

And so you get the skills for understanding the dimension and the level of grain of your data. And thinking about how a lot of times we always use data that's just like past data. My motive is to have people thinking about how we present predictions, how we present what's happening now, and how we situate that in the context of what happened before. And that's all about pitching.

So there's a lots of great interactive activities as part of that. Actually, I've written like two books on online learning. So it's really one of my fields of expertise that I bring to this. There's many other ways and I talk about that in my book too. There's all these free courses that you can take. If you want to brush up, if you want to learn a little bit of a software program, you should definitely do it.

And that's another thing that has really changed about data visualization, not only is there more data, but there's more access to education that allow us to better understand data. And that access is we should definitely take advantage of those resources. We never know how the world will change. Right now, we're in a really great spot. There's a lot of free education out there.

Erik: Data is becoming maybe the most valuable commodity. It's something that every organization from a Facebook through equipment manufacturer now has to have a level of expertise in. And there's really no excuse now: there are great programs, paid programs, free programs out there to help people figure out how to make use of the data that they're gathering. Certainly, we'll put the Amazon link to “Data Visualization Made Simple” in the podcast notes. But is there any other great way for people to reach out with you if they're interested more in learning about you, your work or having a conversation?

Kristen: Twitter's a great way. So definitely tweet me at my handle is just my last name, Sosulski. You can feel free to connect with me on LinkedIn. I tend to post periodic updates of my work there. Twitter is like if you want the daily feed. And of course, I'm really open to email. I'm professor and so it's really in my nature to discuss and have conversation. So I'm really open to that. My email is ks123@nyu.edu.

Erik: Awesome. Well, Kristen, thank you so much for taking the time to walk through data visualization with us today. It's unfortunately not a topic we spent enough time on, but it is incredibly important in the industrial IoT.

Kristen: Erik, this is really a pleasure and super fun. Thank you so much.

Thanks for tuning in to another edition of the industrial IoT spotlight. Don't forget to follow us on Twitter at IoT one HQ and to check out our database of case studies on IoT one.com If you have unique insight or a project deployment story to share, we'd love to feature you on a future edition. Write us at Eric dot valenza at IoT one.com

Erik: Thanks for tuning in to another edition of the industrial IoT spotlight. Don't forget to follow us on Twitter at IotoneHQ, and to check out our database of case studies on IoTONE.com. If you have unique insight or a project deployment story to share, we'd love to feature you on a future edition. Write us at erik.walenza@IoTone.com.

Contact us

Let's talk!
* Required
* Required
* Required
* Invalid email address
By submitting this form, you agree that IoT ONE may contact you with insights and marketing messaging.
No thanks, I don't want to receive any marketing emails from IoT ONE.

Thank you for your message!
We will contact you soon.