Six Components of Layered Graphics

Six Components of Layered Graphics

Layering graphics is a technique found everywhere in data visualization. From coding in R to how Tableau software was designed. By understanding this framework of visual analytics you can take your analysis to the next level. Join Ethan Lang as he walks us through the layers of data visualization and how to apply them.

I’m really excited to be here with you today. My name is Ethan Lang and I’ll be presenting the layered graphics presentation with you today. We’re going to go through a framework of six components that make up the layered graphics. And then we’ll kick off and we’ll go into those in more detail.

Before we get started, just a little bit about myself. So I’m the Associate Director of Analytics Engineering at Playfair Data. I’m also the co-lead of the Veterans Advocacy Tableau User Group. I’m a veteran myself so this is something that’s near and dear to me. But we advocate for veterans and bring light to different issues that veterans are facing through our Tableau user group.

I’m also a member of the Tableau Speaker Bureau and I’m a Tableau Ambassador. So being a part of the speaker bureau, Tableau can call upon me to speak at conferences or Tableau user groups or even to potential clients to explain how I use Tableau in my day to day. And then as a Tableau Ambassador I’m a Tableau user group ambassador. So I work with user group leaders all over the world to share tips and tricks on different things that’s made our Tableau user group successful.

So you guys didn’t come here to hear about me. You guys came here to hear about a much more interesting topic, the six components of layered graphics. So let’s dive into these. So what are the six components of layered graphics?

If we try to translate this framework into Tableau what we’ll find is it very naturally comes to light. In Tableau when you open up Tableau Desktop, the first thing it asks you to do is to connect to a data source. So naturally that’s the first layer within our six components.

Then after you’ve connected the data, you may drag some measures and dimensions onto the view and create some chart. So naturally, again, this is our second layer into this approach or this framework.

After you’ve built your chart you’re going to start adding in aesthetics like color, maybe changing the shape of how the Marks are encoded in your view, Different things like that and those preattentive attributes that are really going to drive that insight or that story you’re trying to tell and bring that to light.

The fourth layer is filtering. And this is where we’re going to start adding in different user interactions where your users can then filter and work with the tool to slice and dice the data to figure out and dive deeper into it to get their own story or find their own insights. The fifth layer is statistics. And this is where we’re going to start adding in advanced analytics to our dashboard to drive home and to bring into light more context, and really start to bring into more context for the end users.

And then last, we’re going to end with theme. And this is taking everything we’ve just built and it’s going to wrap it in a nice well designed dashboard that’s going to drive credibility and adoption of your tools. So these are the six components of layered graphics. And I touched a little bit about how they translate into Tableau. But this framework has been around for a long time.

So a little bit of history behind this idea. The most recent publication that I could find that touches on this framework was written and authored by an author named Hadley Wickham. And he authored a book called ggplot2. And if you’re familiar with R or R code, ggplot2 is a library within R that allows you to visualize data using R code.

So Hadley Wickham published this book. And he actually picked up this idea from another author named Leland Wilkinson. He published this book The Grammar of Graphics in 2005. And it also touches on this framework of these components of layered graphics. And Hadley Wickham even tips his hat to the other author here. The gg in ggplot2 actually stands for the grammar of graphics. So again, he’s kind of tipping his hat to this other author.

But like I mentioned, this idea has been around since we really dawned on us that we had so much data and we needed to initialize it to truly understand it and get insight out of it. And this framework is really what developers have been using since the early 80s at least that’s as far back as this publication here which was the most past publication I could find that touched on this framework. It’s called the Semiology of Graphics. But really this framework has been the building blocks on how we visualize data and how we encode that into a computer to do it for us.

So without further ado, let’s dive into each one of these components in a little bit more detail starting with the first layer, which is data. So like I mentioned earlier, connecting to data in Tableau is extremely easy. They have hundreds of different connectors built in within Tableau. And the moment you open up Tableau Desktop that’s the first thing it’s going to ask you to do is to connect to some sort of data.

So naturally that’s our first layer. Again, this is a very self explanatory step. We’re just simply connecting the data. But there are some fundamentals within this layer that I want to make sure you understand to really get the most out of Tableau and other visualization tools that you may be using.

So let’s dive into some of these fundamentals. The first one being the difference between unstructured data and structured data. So we may have all seen something that looks very, very much like this here where we have dates across the top. We have some measures and dimensions. We have aggregations within our data. We maybe have entire columns or rows that are being used solely as borders between different data points.

Naturally, this is how as a human I would encode data into Excel. I read left to right at least in this culture we do. So naturally I would want my data to be kind of displayed that way. However, if I were to load this in a Tableau we might be able to make some use out of it, but it’s going to be really hard for the tool to get the most out of the visualization if our data is structured in this format.

So we really want it in a structured format, which looks more like this example to the right. You can see that I have some columns across the top and I have all of my data encoded beneath each one of those columns correspondingly. So when I approach this and I’m taking an unstructured data source and I want to structure it, I follow these three rules of thumb. The first rule of thumb is that every column has a column header and that everything beneath that aligns to what that dimension or measure is.

My second rule of thumb is that every row is a unique observation within the data. So you can see here my example, I have Anna, Mark, Max, and Carol. I wouldn’t want to encode Anna twice into my data set. That’s going to cause some duplications and it would really lead to issues further down the line, especially once we start getting into some more of these advanced statistics in that layer and so on.

And my third rule of thumb is that all the values should be encoded with something. And what I mean by that is we don’t want to have any NULLs or NA in our data. If our data is in there and we want to capture it, let’s make sure that we do. And there are times when we may have NULLs or different NAs, but we really want to make sure we’re capturing as much data as we can and filling it out to make it structured and usable.

Now there’s different programs that you can use to take your unstructured data and get it into a structured format. The most popular tools out there right now are Alteryx, Tableau Prep, or even coding with things like SQL, R, Python, or any of those coding languages. All of those you can use to take unstructured data and structure it and you can do it within a repeatable format. Meaning I can take that same unstructured data set that I’m maybe getting on a regular cadence and I can just run it through these programs to structure it and load it into the tools.

Now the second fundamental that I wanted to touch on is this idea between measures and dimensions and how they relate to one another. So a measure is defined as a numeric quantitative value that can be measured. And these would be things like sales, profit quantity, maybe an aggregation like the sum of sales. And then a dimension is a qualitative field that can describe our measures. These would be things like customer name or date or segment. Something like that.

Now how they relate to one another, let’s say I made $2 million in sales and I just knew that figure off the top of my head that measure. Now if I described that measure using my date dimension and I found that I made 1.5 million last year and I’ve made 500,000 this year, I might want to dig into my data and figure out what’s causing that discrepancy. We’ve obviously taken a loss year over year. So now we can start asking those questions. And that’s how we’re getting the most out of our data.

As I go through this presentation, I’ve actually built a supplementary workbook. And I’ll pass it to you guys. It’s on my Tableau Public. But in this workbook I’ll be toggling on each layer as we go. So you can see here, I’ve toggled on the first layer, the data layer. And right now I just have a blank slate. I’ve literally just made a data connection to some stock data, but I haven’t done anything else with it yet.

So moving into the second layer. This is how we want to encode our view. And when I say mark type, what I really mean is choosing a chart type. And when you choose a chart type, Tableau actually makes this extremely easy as well. Right.

We’ve connected to our data. We can go to a sheet and literally just drag a couple measures and dimensions into Tableau and it will automatically build out a chart for us. You’re either going to end up with a bar chart, a line chart, or a scatterplot most likely.

However, Tableau is a extremely versatile tool. So not only does it build those automatically for us, but it gives us the ability to change them directly from the Marks card here. So from the dropdown, if you click on that dropdown within the authoring interface you can choose a different chart type that you want to encode your Marks with. So maybe you had a bar chart and you wanted to encode an area chart, you can make that change directly from here.

Tableau has also built in some very easy drag and drop abilities like the show me tab where you can click on the show me tab in the top right of the authoring interface and simply click on one of these options here. And it will encode those Marks in your view for you depending on the options and the measures and dimensions you have on the View at that time.

However, it goes above and beyond when it comes to making charts. It’s really infinite on what you can do within Tableau and the amount of charts that you can create.

One resource I absolutely love it’s out there on Tableau Public. It was authored by Kevin Flerlage and it’s called the Tableau Chart Catalog. And it has 100 different chart types created by the community. And within this tool you can just scroll through. If you find something that inspires you or something you want to incorporate in your own work, you can click on that and view it within Tableau Public. Even download it and reengineer what that author had done to maybe incorporate it within your own work or to just pick up on new tools and new tactics that you can implement.

Kevin also credits several other authors and tools. The most well known is Visual Vocabulary by Andy Kriebel, The Tableau Resource Guide from Jeff Shaffer, and then the Tableau Cookbook by Josh Weyburn. All of these are incredibly useful tools that you can use to either pick up and learn new chart types or just go out there to get inspired to build something new. So I definitely recommend checking those out.

So as far as chart types today or our second layer, I’ll toggle that on in our supplementary workbook here. And we can see I’ve encoded the data with a line chart in here. So I have the closed price of this different stocks by the day of closing. So you can see that visualized there in the chart.

Now my Third layer is aesthetics. And when it comes to aesthetics, anyone can open up Tableau for the very first time and connect to some sort of data. Anyone can do that. It’s extremely easy. Tableau makes it very easy for even a user that’s never used Tableau before to do that.

Also building a chart type or building out some sort of visualization, it’s extremely easy to do that. All you have to do is just drag some measures and dimensions on the View and you have a bar chart or a line chart. However, encoding and using aesthetics to really drive that insight home or tell that story takes some finesse. So as analysts or developers, this is really where we can start to show our chops and bring value to our organizations.

So I’m going to cover some of the aesthetics that Tableau builds within its tool. One being color. So to me color is probably the most powerful aesthetic that you can incorporate within your dashboard. It’s really going to drive home the insight or highlight specific Marks in your view.

Now we can also see if I add color and I can do so by just dragging a measure dimension onto the color property in the Marks card, I’ll be presented with this menu here. And this is where we can change the color. Maybe I want to edit the colors that are being displayed or use a different color palette. I can also change the opacity to really add in that professional polish to your view.

And I can add effects like borders. Again, just adding some professional polish. We can also add in different line types. Solid lines, dashed lines, and things like that to better incorporate it into our view and make that or drive that insight home, if you will.

The next aesthetic I’ll touch on is shape. And when paired in tandem with colors, again, another great way to draw the eye to specific Marks in your view. Encoding shapes allows you to bring in custom shapes. And we’ll touch on this in the fourth layer. But this is where you can start adding in some different UI like buttons, toggles, and so on.

Next aesthetic is size. And like I said, I keep saying it, when paired in tandem with shape or color it’s a really great way to draw the eye to specific Marks in your view. My only complaint with size is if you’re trying to adjust the size, you’re just left with the slider. So if you’re looking for precision Tableau doesn’t make that very easy for us as developers. But maybe this is something they can address in the future.

Next is axis. And this often gets overlooked as an aesthetic. But for me I do find it as an aesthetic that you can add and tweak to really drive that professional polish within the tool. So within the Edit axis menu here, we can change the range of our axis. We can change the scale. We can reverse it or add a logarithmic algorithm to normalize the data directly from this authoring interface here.

We can edit the titles. We can also edit the tick Marks. So if you’re looking at incorporating an axis but maybe you’re just trying to take less of those tick Marks off the view and make it a little bit more minimal, you can do so directly from this menu here. Another key that I always try to do is incorporate some sort of instructions using my title and subtitle within the axis menu.

So maybe we have a scatterplot and we want to incorporate some instructions on our view for our end users that says something like, hover over a mark to get more data in the tooltip. You can simply do that very minimal and add that directly into your view right here from the Edit axis menu. Now the next few I’ll touch on very briefly. Not to say that they’re not important because they definitely are, but they’re a little more self explanatory than the others.

The first is labels. And this is simply just adding text labels onto your Marks and your view. You can do this by, again, just dragging something onto the label property there in the Marks card. And what I would say about labels, you definitely don’t want to go over the top with it and label every single mark in the view.

That might muddy things up. But if you click on the property here you’ll be presented with options to maybe add a label to the line end or the line start or the min or max or both. That way you can give the user some more context without muddying up the entire view with labels.

Next is details. So if you drop anything to the details property it’s just going to change the level of detail in your view. Tooltip is a really good feature to change and incorporate within your dashboards. I never recommend going with the default tooltips that Tableau gives you right out of the box. I will typically go in there and add some different formatting to just make it a little bit more polished up.

You can also incorporate viz in tooltip. It’s a feature that’s near and dear to me. It feels like it just came out, but that was years ago now. But you can add in a vis in tooltip to, again, drive more context for your users. And the last aesthetic I’ll touch on is angle.

And before anyone tries to jump on me about selling them into a pie chart, you can use angle in a lot of different ways other than pies. Especially when you get into some of those more advanced chart types that I mentioned earlier. So it’s definitely a very useful aesthetic that you can apply and play around with. All right.

So my third layer I’ll toggle that on here. And we can see I’ve added color to my view now. So I have colored the closing price if it was positive versus the prior day I’ve colored it green. If it’s down from the prior day I’ve colored it red on this line chart.

So the fourth layer is filtering. And like I said earlier, this is where we can start diving in and allowing the user to interact with our tool and slice and dice the data. Come up with their own insights and make decisions on the data. Within Tableau there’s actually eight different types of filters. So I’ll be taking us through the order of operations and just covering briefly on each one of those, how they work, and what’s the best use case for each one.

So let’s dive in here. So processing first starts at the top and goes down this list. So we can see the very first thing that Tableau processes is actually a filter. As a matter of fact, the first three are filters alone. So the first filter to be processed is called an extract filter. And then the next one is a data source filter.

I’ll cover both of them because they are related in some way. They’re both implemented directly from the data source tab. So within the authoring interface, if you go to the data source tab it’s at the bottom left of your screen. You can apply a filter from the top right of that. So if you look at my screenshot to the right, you’ll see something that says filters to the far right. And then there’s a button that says Add in blue font. If you click that and add a filter there, that’s actually a data source filter.

Now if I change my connection to an extract, what will happen is right next to that extract radio button you’ll see another filters get popped up there and a button that says Add again. So you’ll have two add labels. If you click on that second add label that it pops up when you create an extract of your data, what will happen is you’ll create an extract filter. Now the biggest difference that I can find between these two an extract filter will actually limit the amount of data that’s processed when you package your workbook and create that extract.

So let’s say I had 20,000 observations or rows within my data. If I created an extract filter that took that to 5,000, I would literally only have those 5,000 rows of data within my workbook packaged within the extract.

Now if I run a data source filter on that same scenario, let’s say I only had the data source filter applied. These are two mutually exclusive examples here. I create a data source filter and I add that to my view and I apply the exact same logic. The 20,000 rows of data will still be in my workbook. It will still be there. But everything being processed will be processed on the 5,000 rows of data moving forward once we start building things in our view.

It’s pretty confusing, but there are advantages there. Especially with the extract filter when you’re creating extracts and maybe sending that packaged workbook off. If you create that extract filter, again, it’s limiting those number of rows. If you publish that out there, the underlying data is only going to have the 5,000 rows. Both of these have their advantages. I definitely recommend looking them both up. They can also be used in tandem. So you can have an extract filter and a data source filter applied at the same time.

Now the next filter to be processed is called a context filter. Context filters are extremely important. And you can use them to get this really nice cascading filter down through your filters within Tableau. You’ll see my screenshot here to the right. I have two things on my filter shelf. I have a category and subcategory.

Notice that category is in gray. This means that it is a context filter. To make a dimension filter a context filter if you add it to the filter shelf you can right click on the pill. And there’ll be an option that says Add to context. Selecting that will create a context filter and move it into context.

The next three to be processed I’ll highlight at the same time. They’re all implemented the same way, but they have slightly different logic. And as you can see they’re processed a little bit different. So if you add any dimension onto your filter shelf, you’ll be creating a dimension filter. And you’ll be presented with this menu here to the right.

So in my screenshot you’ll see four tabs across the top of that menu. We have general, wild card, condition, and top. If I navigate to the Condition tab and I apply some conditional logic to filter my view, that will create a condition filter. And as you can see, it’s process before dimension filter.

Same with top end. If I navigate to the tab labeled top and create some logic that’s going to filter down to the top end filters, that will create a top end filter. If I stay on general and I just select some dimension members, that will create a dimension filter.

The next filter is a measures filter. And if you add any measure to the measure shelf you can create a measures filter. And you can use this let’s say a good use case would be I want to look at everything with $10 in sales or more. You could apply logic like that with a measures filter and filter your view down.

And then the last filter to be processed is called a table calculation filter. This is applied very similar to a measures filter. However, in my screenshot notice the very bottom pill has that delta sign to the right of the pill. This denotes it as a table calculation. And applying a table calculation, like I said, is very similar to a measures filter except it’s being applied on a table calc logic.

So that was a whole lot of filters. But to drive it home, I’ve created this view here. And really when we add filters to our view we’re not trying to just add a bunch of filters and get down to something. We really want to incorporate those filters to allow our users to then interact with our tools.

So you can see here my screenshot. I’ve added a filter a few of them here. And I’ve allowed the user to select a date range as well as select a different stock. So within this tool now the users can select stock A and maybe get an idea of what stock A is doing. And then select stock B and then make some sort of decision on that. Really that’s what applying filters leads to is that user interaction. And that’s why they’re important.

So again, I’ve toggled on that fourth layer. I’ve added in some filters. And I’ve applied that to my dashboard allowing the user to filter around these different stocks.

So my fifth layer is statistics. And with statistics adding any kind of advanced analytics to your tool. Obviously, you have to be careful with what that brings to the table, especially for non statistical individuals or stakeholders. You don’t want to overwhelm them with a bunch of stats and summaries and all this stuff. But you do want to incorporate that stuff to visually appeal to them and show them what the statistics is trying to tell them and convey.

So let me show you how you can do that within Tableau. If you’re on the authoring interface, you’ll have two panes. You’ll have the Data pane and then you’ll have an Analytics pane. If you have something on the View, you can toggle to the Analytics pane and you can literally drag and drop these models onto your view. And what you’ll be presented with is what you’ll see here to the right of the Analytics pane where it says Add a distribution band.

It will allow you to add that statistic to the x-axis or y-axis, which would be denoted by the column and rows. You can add it to different level of detail. So you can add it across the entire table, specific panes, or individual cells.

And all you have to do is grab one of these options from the Analytics pane and just drop it where you want. So in my example here, I’ve highlighted average line. If I were to drag the average line and drop it on to table, it will add an average line going across my y-axis and my x-axis creating almost a quadrant view. This is very useful if I’m working with a scatterplot, again, so you can see and navigate between those different quadrants. And maybe come up with some sort of segmentation or make some decisions based on that.

The other one is an average line with a 95% confidence interval. And this is where we can start to get into some of the different modeling. But a 95% confidence interval will add some banding that will show you 95% if everything kind of stays the same. We’ll see that the average falls within this confidence interval 95% of the time.

Another great one to use or incorporate is the median with 95% confidence. I like to incorporate this especially in tandem with an average line. And what that will do is it will show you if your data is normalized or not. If your data follows a normal pattern and it’s distributed normally, the average and median are going to be pretty close to one another. However, if you have any kind of skewness in your data, whether it’s skewed left or right, you’ll see those average and median lines kind of start to navigate away from one another.

So they’re pretty useful when applied in tandem. The next few I’ll touch on are some of the different advanced statistical models that you can put within Tableau. And what I love about this is it takes something extremely advanced. We could talk about modeling all day, probably all year if I tried to explain some of this stuff. However, Tableau takes that stuff and it makes it very easy for us to implement and understand.

So for instance, the first one I’ll cover is trend lines. And adding a trend line is exactly what it sounds like. It’s simply a line that will show you the trend over time of your data. So we’re presented with these options here. And again, I love the fact that Tableau has taken these advanced models and broken them down to where we can see very visually if our data follows a linear pattern moving from the bottom left to the top right. Let’s try a linear line. If it kind of goes up and down, maybe we’ll try a polynomial model instead that will fit the data better.

So again, Tableau makes it extremely easier for us to interpret and add these things in on the fly to drive more context home. So we have the ability to add linear, logarithmic, exponential, and polynomial, and power trend lines to our view by simply dragging trend line and dropping it on one of these models. And then it takes it a step further.

For those of you that are a little bit more statistical savvy, you can actually hover over it or right-click on any one of these trend lines and view the underlying advanced summary statistics for these models. So you can start pulling together your R squared, your coefficients, and maybe making some descriptive contextual decisions about your data.

The next model I’ll touch on is the forecasting model within Tableau. It’s actually a pretty advanced model. They incorporate a model called exponential smoothing with their forecasting. And again, what I love about this is it makes it easy even for people that aren’t statistical savvy. If I drag forecast onto my view, if there’s not enough data there to create accurate forecast, what Tableau will do is just draw a line across the average moving forward.

So if we’re trying to forecast out a few years, if there’s not enough data there you’ll just end up with an average line telling you maybe we should get our data more granular or maybe there’s not enough data to come up with an accurate model. So in those cases, again, like I said, you could maybe take it from year to week or year to month to week to day even to create a better forecast that will be more accurate. And you’ll probably start getting more results.

Again, for the more statistical savvy folks, you can right-click on these. You can change the number of periods that you wanted to forecast forward. You can start to view different summary statistics like AIC and all of those things. So again, it kind of is that line between people that aren’t statistical savvy versus who people that are, and teaming up and working well together to visualize this data and make sense of it.

The last model I’ll cover is the clustering model and within Tableau they use a model called K means. And what that will do is create different clusters. Again, it’s super easy to change the number of clusters you want to create. And this is a great way to start making sense of maybe you’re trying to segment out your customers and make decisions on them. This is a great model to incorporate that and visualize that so then you can start creating those segments.

Now Tableau, like I mentioned, is also a very versatile tool. If you want to do more advanced statistics or use different types of models they allow you to do that. All you have to do is go and click on the Help button at the very top. And I created this screenshot I think a couple versions ago so it might be slightly different. But overall, it’s going to be about the same.

You’ll navigate to Settings and Performance. And there should be an option that says Manage Analytics Extension Connection. From there you can connect to R server or Python. And you can write your R or Python directly within Tableau. It will pass it over to those other external tools, process it, and bring it back into the tool. So again, if you’re looking to for more advanced modeling and you want to visualize that in Tableau for your end users, Tableau gives you an option to do that.

So my fifth layer I’ll toggle that on. You can see I’ve incorporated a couple of things here into my view. I’ve added a trend line and I’ve also added some forecasting. So both of those were brought in using the Analytics pane. And you can see how that adds in some more context for the end user. So if we’re trying to make decisions on this data, this is a great way to, again, add some more context so the users can make an informed decision.

Now the last layer is theme. And thus far, everything I’ve covered Tableau makes it extremely easy for us to implement and build within the tool itself. Everything from connecting to data, to visualizing it, incorporating different aesthetics, even bringing in advanced modeling into our views. Tableau makes that extremely easy.

However, when it comes to simple design elements like maybe adding in some curved containers or bringing in some drop shadowing effect to add some depth to our view, Tableau doesn’t do that at all within the tool itself. It’s very hard to do that. There are different hacks out there where you can add in some drop shadowing with different containers and whatnot. But Tableau they have not perfected this design element within the tool itself.

So when it comes to theme, I look outside of Tableau to other external tools. These two are probably my favorite the ones that I use the most often. One is Figma and that is the logo to the left. And now that I’ve realized I haven’t encoded these correctly. I should add some labels. But Figma is on the left. Adobe Xd is on the right.

The advantages of both of these. Figma is a free tool. You can go there to Figma.com, create an account today, and start building in incorporating some of these design elements directly within the tool. Again, that’s the major benefit is it’s free. It’s very easy to use and learn. And there’s a lot of stuff out there different forms that you can pick up on to find out more tricks and stuff like that.

Adobe Xd, however, is a subscribed software. So you have to buy in to Adobe Creative Suite. And that comes with Adobe Xd. Essentially does the same thing Figma will do, but it comes with a little bit more bells and whistles. So within the Creative Suite, you get access to things like Photoshop, or Illustrator, different Adobe tools that could also bring value to you or your organization.

So there’s definitely pros and cons of each tool. I’d recommend checking them out before making a decision. But all in all, what we want to do is go into these tools and design a theme. And a theme is really just taking that view or that visualization that we’ve built and wrapping it into a nice well designed dashboard.

So you can see here within our view, I’ve toggled on theme. And I’ve created some architecture around my view that adds that professional polish. What this is going to do is drive credibility of your tools. It will also drive adoption. I always relate this to the cars on a car lot. If I went to the car lot and I saw an old rusty beat up car and then I saw right next to it, a Lamborghini. Even if the Lamborghini didn’t have an engine in it, I would probably walk over there and check it out. However, the rusty beat up car might be the last thing I look at.

All right. So I’m going to try to transition over to my supplemental workbook I’ve built in Tableau Public. Let me see if I can make this happen.

All right. Here we are.

So this is in Tableau Public. And this is essentially the same view. And I’ll refresh it here so it doesn’t do it for me. This is the same view I’ve been walking us through and toggling on as we’ve gone through the presentation.

You can see here I have the six components of graphics here. And what this allows us to do is toggle these things on and off to see how they work with one another. So what do I mean by that?

So for instance, if I toggled off my aesthetics layer, we’ll see some different layers be affected by that. So if I remove aesthetics, which was color in this case, you’ll see that I’ve now lost some of the forecasting that I had built in my statistics layer. The reason that is is we need color to differentiate what’s being forecasted and what’s actual. And that’s what Tableau uses is that color aesthetic. So you can see by toggling off that we actually lose some of the functionality in these other layers.

Again, if I toggle off theme. We can see I still have an awesome visualization here. However, I’ve lost that really professionally polished dashboard that I’ve built around this workbook. So you can see I’ve now lost that credibility. I’ve lost some headers, some different titles, and things like that.

If we toggle off filters, you can see that I’ve now lost the ability to play and interact with this. So now I’m just presented with the data. This might not always be the bad case. However, we do want to allow our users to interact with the tools that we build and make decisions or the best decisions based on that.

So I’ll send this out. This is available. It’s in my Tableau Public portfolio. So if you go there you can check it out and download it and see how it ticks and how I built it and everything else. Now let me try to hop back over to this without busting nothing. All right. So perfect. All right. So now I’m going to cover just one moment. Sorry, folks.

All right there we go. So with that, I’ll just end with presenting you guys with some resources that I definitely recommend checking out if you haven’t already to upskill yourself or maybe even just to find some inspiration. But one tool I definitely recommend is Tableau Public. When I first started getting involved with Tableau Public, I was very nervous to get my work out there in front of different individuals.

But what I found is Tableau Public is way more than just building something and publishing it out there for everyone to see. Tableau Public for me is a place I go to find inspiration if I’m trying to build a new tool. I maybe stop through Tableau Public to just peruse through and see what different authors have done. Maybe something will catch my eye.

It’s also in a way one of the world’s largest data repositories. So if you’re interested in something particular, I guarantee you could go to Tableau Public, search for that, and you’ll probably find an author that’s built a data source and connected to some data and visualized it within Tableau Public. So it’s a pretty cool tool to just go out there and look for data, look for inspiration, and obviously build your own portfolio if you’re comfortable with that.

Other resources that you can take advantage of and continue learning with Playfair Data. Obviously all of you guys are Playfair+ members and you’ve probably connected to our blogs. But we also do offer training and consulting. So if you or your organization ever wanted to partner up with us, definitely feel free to reach out. And we’ll start looking into that.

Other resources out there in the community. There’s tons of blogs authored by Tableau, by other Datavis enthusiasts. These are some of the ones that I definitely recommend. Tableau does a monthly blog and a weekly blog. Other resources like the Flerlage Twins. I referenced them several times. I love their blog and the work they do.

And then there’s also a workbook out there. I think Andy Cotgreave published it. But there’s like 1,000 blogs that he’s tracking within this workbook and there’s links to all of them directly from this workbook. So another cool resource to check out.

But all in there, I cannot see the chat. So if there are any questions, feel free to post them in the chat and someone will come and bring them to me. But if you want to connect with me in more detail, here’s my email address. It’s [email protected]. My Twitter handle is @thedataveteran. And then on my LinkedIn URL here is just the linkedin.com/in/ethan-lang.

So with that, I will say thank you guys so much for joining us today for our second quarterly webinar. We really appreciate taking the time out of your day. Hopefully this was valuable. Again, if you guys do have any further questions, feel free to reach out to me with my contact information there. Happy to answer all of that and have a wonderful day.