Southern Africa Tableau User Group 18 Jun 2026
Mapmaking with Tableau: Crafting Eye-Catching Maps with Dennis Kao
52 min 7,933 words transcribed
Summary
The Southern Africa Tableau User Group's June 2026 meeting, Mapmaking with Tableau: Crafting Eye-Catching Maps, hosted by Suzanne Edlmann, who leads the group with Craig Andrew and Tienie de Klerk. Dennis Kao, a social work professor at Carleton University and co-lead of the Data + Maps TUG, explains why after 15 years of GIS work Tableau is his favourite tool for combining map making, analytics and design freedom, then covers the basics: reference versus thematic maps, maps as layers of points, lines and polygons, vector versus raster data, and how map layers made Tableau far more GIS-like in 2020. Building live a map of South Africa's ten largest cities from free data (the Humanitarian Data Exchange, Open Africa and OpenStreetMap), he starts with geographic roles and graduated circles, swaps Tableau's map fonts for Arial, adds cities as separate label layers so manual label positions survive, and explains why he turns off Mapbox backgrounds that can't be restyled. He then brings in shapefiles to build his own base map: a country geometry with the OUTLINE spatial function for a thick border, lighter provincial boundaries, an extent map of Africa in place of the neighbouring countries, and trial layers of water and major roads. His closing tips are to keep it simple, never accept defaults and take advantage of map layers; in the Q&A he describes preparing spatial data in QGIS before visualising it in Tableau, and says a good eye for design comes with practice.
Listen to this TUG
Prefer to listen? This meeting is also an episode of the Tableau TUG recordings podcast.
Key points
- Dennis Kao: GIS software is best at spatial data, but Tableau combines map making, analytics and a blank design canvas better than any other tool.
- A reference map shows everything and lets the reader find the story; a thematic map focuses on one phenomenon, so you can leave most detail out.
- Since map layers arrived in 2020 you can stack as many layers as you like without joining the data, much as maps are built in GIS.
- Spatial functions such as OUTLINE and BUFFER only work on spatial files like shapefiles and GeoJSON, not on geographic roles, so bring in spatial data.
- There is plenty of free spatial data: the Humanitarian Data Exchange, Open Africa and OpenStreetMap supplied everything in the demo.
- Hand-placed map labels can reset when the view changes; putting groups of cities on separate label layers keeps their positions.
- Mapbox backgrounds can't be restyled inside Tableau, so for full control build your own base map from your own layers.
- Use border weight to show hierarchy: a thick country outline and lighter provincial boundaries read at a glance.
- Drop context your audience doesn't need, such as neighbouring countries and oceans, and add a small extent map to show where the area sits.
- His workflow: clean, clip and prepare spatial data in QGIS, then visualise and publish it interactively in Tableau.
- Keep it simple, don't accept the defaults, and take advantage of map layers; a good design eye develops with practice.
Jump to a chapter (17)
- 0:00 Dennis Kao: mapmaking with Tableau
- 1:49 Why Tableau for maps
- 3:33 Reference and thematic maps
- 6:59 Layers, vector and raster data
- 7:42 Map layers and spatial files in Tableau
- 11:16 The approach and the free data sources
- 15:00 A quick map with geographic roles
- 20:37 Fonts, labels and label layers
- 24:26 The trouble with Mapbox backgrounds
- 26:20 Outlining the country
- 29:43 Building your own base map from shapefiles
- 32:58 The OUTLINE function and border weights
- 37:31 Dropping context and adding an extent map
- 42:27 Trying water and road layers
- 47:22 Tips and the Data + Maps TUG
- 48:30 Q&A: Tableau versus GIS
- 50:41 Q&A: developing a good eye
Transcript
0:01 Dennis Kao: Well, hello everybody. My name is Dennis Kao.
0:04 Uh, I'm really pleased and honored to have been invited to s- present.
0:09 Uh, k- a couple things that are just my favorite things in the world is, uh, uh, I love to… I love maps.
0:18 I love creating maps and designing maps.
0:21 Uh, and I do so both professionally as well as personally.
0:26 And, um, you know, a more recent kind of love of mine is Tableau, which I started back in November 2020.
0:36 Uh, this is a little bit about myself. I'm a social work professor at Carleton University. I'm, I'm originally from Los Angeles, but we moved here about eight years ago, uh, when I took the job, uh, joining the faculty at Carleton University. I've been doing statistical ana- statistical analysis for about 15… over 15 years, GIS mapping and spatial analysis since 2020.
1:00 Uh, but recently, um, discovered Tableau, uh, about six years ago, or five and a half years ago, and just, you know, it kind of just, um, I fell in love with the software and, um, uh, I use it pretty regularly in my, both my professional and personal, uh, um, map making. And so,
1:25 you know, I, I think it's, it's really… You know, I use kind of all sorts of software. Stata, uh, R, SPSS in terms of s- statistical analysis, ArcGIS, QGIS, and things like that. Um, but Tableau has kind of really turned kind of my career around, and it's kind of re- re- kind of re- reignited kind of a passion that I might have with data visualization and mapping.
1:50 And so you'll kind of hopefully see that, uh, kind of passion come out as I present about map making in Tableau.
1:59 But why Tableau? So I've been u- I've been using GIS software. I, I was trained in ArcGIS, um, during my PhD program, and I, I since y- I've turned to QGIS because it's free, and it offers mu- you know, a, a lot of the functionality that you will find in ArcGIS Pro.
2:22 Um, but why Tableau?
2:24 I think Tableau, in my mind, uh, where GIS software is really good at dealing with spatial data, it's specialized software, uh, Tableau isn't, right? It wasn't built to deal with spatial analysis, uh, spatial data. But you know, of course it can handle spatial data.
2:42 Uh, GIS software is also specialized, you know, when it comes to making d- digital maps. But in my mind, in my humble opinion, after kind of 15 or so years of doing map making and now turning to more data visualization, I think Tableau really is the best tool that combines map making, data analytics, and offers the design kind of freedom, right? Like that blank canvas to kind of create your dashboards or your infographics.
3:15 Um, kind of all that combined, I think really puts Tableau o- over any other tool in my mind.
3:23 And so that's after… And, and that, obviously that's my opinion and, and why I've kind of now use Tableau pretty regularly in my work.
3:37 So before I go into kind of map design and map making, I thought I would just go through a couple things that, you know, just kind of basics about maps, uh, for those of you who aren't familiar with maps.
3:49 Um, maps are kind of symbolic, condensed representations of reality. And so when you look at a map, even at on Google Maps, uh, or a paper map, or any map image that you may come across, w- the effort, it's basically an effort on the part of the map designer to take reality and symbolize it using kind of pictures, images, symbols, things like that.
4:19 Um, there's basically two types of maps. There's the reference map, which are, you know, if you think about Google Maps, it's basically put everything on a map.
4:28 Um, everything's all equal. There's really no kind of… It's not guided by any topic, and the reader is really the one that kind of drives the story and uses the map as, uh, a way to kind of weave a story, whether it's to f- just find directions to the, the nearest hotel or Starbucks, or to look at kind of what, how far different places are.
4:52 Um, a reference map tries to put out as much information as possible. I think it's akin to like kind of a dashboard where you just kind of visualize the data as m- as best you can, and then the reader is the one who kind of figures out and plays with it, interacts with it to, to kind of find the story in the data, right? So I think it's akin, a reference map is probably akin to a dashboard in the purest sense.
5:20 Whereas a thematic map, which I think is what we kind of tend to focus on as data vis… I- in kind of the data analytics world, is w- we focus on thematic maps, which really key in on a specific phenomenon, a specific topic. It's designed, you know, we design…
5:40 The designer is driving the story, right? And so, and it's, and, and, and because we're driving the story and we're focusing on a phenomenon, there's a lot less detail in the thematic map. And hopefully what, what I, w- what I present, you know, why I'm presenting these two types of maps is hopefully it allows you to have, feel more f- like a little bit more freedom- To be flexible, to not include as much detail as, as possible, but that freedom to say, "Oh, let me just key on the very specific things I'm interested in focusing on, and everything else can…
6:18 You know, I provide enough context, but I'm, I'm really just focusing on the specific phenomenon that I wanna focus on in the map." Right? And, and an example of a thematic map might be this, where I… This is a quick map that I put together a few years ago on the rivers in the United States.
6:35 Um, there's no other data. It's just one kind of layer essentially.
6:41 Um, there's no streets, there's no cities, there's no nothing. Um, but it was my way to kind of represent the United States, right? And, and, and, and you can still see the kind of the, the outline of the United States, um, but just based solely on kind of the rivers and the river basins.
7:02 So when you think about digital maps or a GIS map, uh, y- you know, you gotta start thinking about, um, you're gonna s- what you're seeing in Google Maps, for example, is just a bunch of layers, right? You might have a layer of points, which might be Starbucks; a layer of lines, which might be roads; a layer of polygons, which might be parks, et cetera, and they're all kind of overlaid on top of each other in some kind of way and, and, and hopefully in a way that is…
7:29 makes the map easily readable and things like that, and clear.
7:34 There's two types of spatial data, and in the Tableau world, we primarily operate with vec- vector data, which are points, lines, and polygons. But there's another type of data called rasters, which are essentially kind of, it's just a grid with kind of each cell, each cell has a specific value.
7:54 Um, and, and so if you imagine weather maps and think, or satellite maps, those are kind of examples of raster data.
8:06 Now, the cool thing about map making in Tableau, ever since 2020, um, when they introduced map layers, Tableau has become very GIS-like, right? Uh, so I think prior to that, in order to create a map and to have kind of, you were almost wedded to just two layers, right? Uh, 'cause you had to use, um, dual-axis kind of functionality in Tableau. But now with map layers, you no longer have to connect the data to each other.
8:35 You can just plop in the data, and then you can create layer upon layer upon layer, which is, I think is very akin to how, if you are familiar with working in the GIS world, that's how maps are created, right?
8:47 Um, just layer upon layer, right?
8:50 Um, I think there's… it's great that, you know, Tableau also makes it very easy to create maps and, and, and you can create a quick map using, like if you had a map, if you have data of cities, for example, you can use the built-in geographic role functionality in Tableau to quickly map those cities, right?
9:12 Um, Tableau can also handle various spatial file types, so shapefiles, geodatabases, GeoJSONs, et cetera.
9:21 I primarily, I've been trained on shapefiles, which are kind of antiquated at this point, but, um, that's kind of my, how I was trained, so I'm still most comfortable with using shapefiles.
9:33 But when you start bringing in kind of separate shape- spatial data, like a shapefile or geodatabases and things like that, and move away from built, uh, using the built-in geographic roles,
9:48 that's where Tableau has been introducing a bunch of spatial functions. For example, outline, buffer, and things like that. You need to be able to, you need to… Those can only be done with kind of spatial files, right? You can't do that with the, the cities, uh, that are w- you can't do that with geographic roles.
10:08 Um, and so, so if… You know, I will often encourage people to bring in spatial data. There's a lot of free data out there.
10:17 Um, I'm not sure how your organizations work, but all the spatial functions like buffers, outlines, things like that that might be kind of a value, um, and have been coming out in the past couple years, those can only be done if you b- once you start bringing in kind of spatial files. And so, you know, I… obviously, that's kind of one advantage, um, if, if you wanted to use those spatial functions.
10:44 Okay, and so that's enough for basics, you know? You know, um, because this is kind of an open… This is not a webinar. Like, if you have any questions, you know, please feel free to, to chime in. Um, I mean, maybe Suzanne, if you can kind of keep me… Like, you know, I'm not checking the chat at all, but if you, if there are any questions that come up, you know, in the meantime, I'm happy to, to entertain them throughout the presentation as well.
11:11 All right.
11:13 Okay.
11:14 So what I'm gonna be focusing on is less on Tableau. So I'm still probably relatively new, although six years, can I, can I say I'm a Tableau newbie at this point?
11:25 Um,
11:27 you know, I've been using Tableau for six years, um, but I think I've become very good at, like, very specific things, right? Um, there's a lot of Tableau that, you know, I still struggle with, and so there's very specific things that I'm, I think I'm pretty good at. And the one thing that I think I'm pretty good at is kind of m- map making, um, and which is less kind of maybe Tableau functionality, but kind of more broader spatial thinking and…
11:57 right? And so my… And, and, and it's kind of me applying m- what I learned using GIS for 15 years and, and kind of- A-adapting that into Tableau Space, right? And so, um, so some of the things that I do are, are probably kind of rudimentary in nature.
12:15 Um, but I'm gonna- what I'm gonna key in on is some of the kinda little things that you can do to maybe ma- make your maps a little bit more effective, clearer maybe, a little bit maybe pop out maybe.
12:28 Um, and so th- that's what I'll be talking about, um, once we get into Tableau.
12:34 Okay?
12:35 All right. There's a few data sets that I'm gonna be using that I pulled from various places.
12:41 Um, one kind of caveat before I m- move forward is I often, because I, I do mapping in my professional wo- professional work, I often am creating maps and doing spatial analysis in places that I'm pretty familiar with.
12:58 Um, I've collected data from these places. I've been to these places. I've driven around these places. I'm interested, like my kind of like I live in these places. I'm pr- I'm, you know, and my professional interests are in these places. So for example, I, when I lived in Houston, we did a mapping project where we, we, we, we interviewed older adults and se- uh, s- uh, well, seniors, um, on kind of where they go t- uh, for their healthcare, and then we kind of put that together in a map, right?
13:31 And so w- I have the context. And so in, in this demo and as I move forward, um, I don't really have that context. Um, being raised in the American educational system, um, geography wasn't kind of a strong suit, i- uh, being raised in that, in that educational system, and so I'm not very familiar with the, with South Africa in general. Not familiar with the cities.
14:01 Uh, you know, I, you know, I'm aware of some of the bigger cities and things like that. Um, but so what I'm doing is I, I map… I'm creating this map and, and there's not gonna be a lot of context unfortunately, so I apologize that- for that, uh, in advance.
14:16 Right. So I have a few data sets I want to, um, and, and the basic gist will be to map the major cities in South Africa. I'm just trying to keep it simple and kind of maybe go through some of the steps that hopefully, um, will be helpful for you. I have some administrative boundaries that I pulled from the Humanitarian Data Exchange, uh, Africa countries in a shapefile from Open Africa, and then I have some roads and water, uh, polygons that I got from OpenStreetMap.
14:44 Okay. And so there's a lot, and all this data is free, kind of somewhere in the internet. Um, and so hopefully that's, uh, kind of a tip that, you know, d- if, if you're looking for some data, there's just a ton out there.
14:55 Um, I don't spend a whole lot of, uh, money or resources on data 'cause there's a lot out there.
15:03 Okay, so I'm gonna quickly move to Tableau really quick.
15:09 And what I thought I would do in Tableau, um, is what I w- kind of take, take you kind of step by step.
15:21 And what can we do, like just exclusively with Tableau's built-in functionality, like geographic roles and things like that, and then kind of build up, move us toward, okay, let's start bringing in some spatial data, and how hopefully that enhances our map in some way, and that, and that's, that, that's the goal, right? Ultimately, we wanna create a map that's clear, and Tableau allows us to make quick maps, and sometimes that's, for maybe majority of time, that's just sufficient, right?
15:58 Uh, I… Another caveat is I'm not in your world. Uh, many of you are in data analytics. You had to put out stuff really quickly. Sometimes a simple map that you can build quickly in Tableau is all you need, right?
16:13 I spend a lot more time than the average Tableau user worrying about how thick these lines are, um, where these labels are in a map and, and because that's kind of…
16:25 I'm interested in the map design side, and that's part of my work. And so I, I understand that we're probably not in the same space. But hopefully some of these will translate well to kind of, you know, things that you can do quickly to kinda create kinda more effective, clearer, nicer maps. All right?
16:45 And again, please, you know, raise your hand or chime in if you have any questions as we move, move, as we move forward, right?
16:53 Okay. So I'm gonna just bring in some data right now. And the first thing I'm gonna do is I'm gonna bring in an Excel spreadsheet that has the ci- major cities.
17:07 Plop that into the data view. And you'll see it's just major, the top 10 major cities in South Africa by population.
17:17 Um, I've also included the lat and lon, longitude, la-latitude and longitude coordinates just so that we can play with those later as well. But that's basically the gist of it, top 10 cities based on population in South Africa.
17:32 So let's go to sheet one. And, and already, as many of you are already familiar, because we have a city field and a province field, um, Tableau is a-already automatically recognizing those as geographies, right? And so they already applied the geographic role to them.
17:50 And so we can simply drag this into the field.
17:54 And of course, it doesn't work for me 'cause I'm based in Canada. So let me just quickly
18:04 change that.
18:06 And I believe this one-
18:11 And, uh, I, I probably butchered this if I tried to pronounce it. Is, is it Gqeberha?
18:17 I'm sorry.
18:18 Um, I believe that is Port… used to be Port Elizabeth. Is that correct?
18:29 There we go.
18:30 So I'm gonna put that in.
18:32 All right, so now it quickly maps it, right?
18:36 Um, and I'm gonna make these circles,
18:41 and there you go, right? And then, you know, in, in, in Tableau, you have these kind of Mapbox background layers, which I have always turned them off, but let me just add a few.
18:57 And then maybe we want the country regions, province regions, and, and that's about it, and I'll leave it at that, okay? Let me adjust the,
19:13 what we're looking at here.
19:16 Okay.
19:18 And then we can, you know, obviously, um, I can create a graduated
19:25 s- circle map based on the population.
19:35 Let me change the color to that, maybe to… and then op- opacity, I'll decrease to 60, and then outline, I'll make a kind of a white like that, so it kind of stands out a little bit.
19:48 Okay.
19:49 And we can drill down a little bit more as well.
19:53 Okay.
19:57 Adjust the size. And then what I can do, because w- we ha- we, uh, Tableau h- now has map layers, what I can do is, okay, s- you know, with the graduated cylinders, uh, the, the size of the circle represents a population, but you don't really know, for example, where the city is. And so maybe I'll just add another, uh, city as another map layer.
20:22 Make those also circles, but not by color, by detail.
20:28 Make those black,
20:33 and then, um, kind of adjust the circle t- to look like that. And then what I'll do is I'll label the circles instead of the graduated, the, the little points instead of the graduated cylinders, uh, graduated circles.
20:54 And then I'll just, uh, the labels like so. Okay? And so that's something, you know, I, you know, I'm kind of going through this quickly, but maybe this would, this would, uh, this kind of is one way to kind of map this out. One thing that I have kind of just based on my own personal preference is I don't, I'm not, I'm not really fond of Tableau fonts on maps specifically, and so I'm just gonna change the font.
21:22 Um, you know, I use a variety of fonts based on the kind of the theme or things like that, but I'm just gonna s- go to Arial.
21:29 And I think one thing you'll notice that Arial just seems a lot more clearer, maybe darker on the map as well.
21:36 Um, let me see. Is there anything else I need to do here?
21:41 Okay, now that works. I'll make, I'll probably make circles a little bit bigger and then make the font a little bit, uh, to 11, like so.
21:51 Okay. And, and, you know, that's, that's a quick map, and if I was gonna plop this into a dashboard really quick.
22:06 There you go.
22:07 Now, one thing I don't nec- I don't really like about Tableau is, um, some, oftentimes, sometimes what you see on the worksheet is not the same as what you see, especially with maps, not what you s- not what you see on, um, the dashboard. And so I might adjust this a little bit just to, um, make sure that we are capturing what we're, what we need to capture.
22:34 So oftentimes I'll just shrink it and then resize it, like so.
22:40 Okay.
22:43 And then maybe move it around a little bit. Okay. All right.
22:46 So that, you know, i- in essence could be a functional map, right? It, it could work. Um, maybe I'll move the, move this label a little bit here.
22:56 Um, like I might adjust the labels once this is all set. The only problem with manually changing these labels is that if you do anything, like you turn them off or you, you, you add data, you know, whatever you do, sometimes these marks, these mark labels get reset, right? And so sometimes it might be helpful to, to, to hard code these. And so one, one thing that I might do is add the labels kind of separately.
23:23 Like some labels I might want on the left side, so I might add those specific cities as one layer.
23:31 Um, and so I can adjust the labels that way. And then for the other cities, I can adjust it that way. And so that's one way of kind of like you can take advantage of map layers, is you can a- add as many layers as you want. It doesn't really add anything, right? 'Cause you're not adding new data, right?
23:50 Okay.
23:51 All right. And so, you know, I, you know, I, and I'll continue kind of playing with this 'cause I can't really see the other circles here and so maybe I'll just adjust the circles a little bit like so.
24:04 That, and that's the problem, right? Like I don't… like the sheet doesn't look exactly like the dashboard sometimes, but I think that looks a little bit better.
24:12 Maybe I'll make this
24:17 Darker color? Maybe that would… No.
24:21 Like, I would have to kind of play with it a little bit, right?
24:24 Okay.
24:25 All right. So far, so good. Okay.
24:27 Now, one thing that I don't like about the map bo- using Mapbox layers, which I think are great just generally 'cause they provide some context. You know, obviously, if you wanted to show something satellite, that, that's, you know, another way you can do it, um, which often is, you know, depending on the context, might be helpful.
24:48 Um, I can't do anything to this, to this background layer besides adding them and s- and, and, and, and removing them, right? I can't adjust this line.
25:02 Like, like, let's say I wanted to make the outline of South Africa darker.
25:06 I can't do that, right?
25:09 Um, without adding additional information, right? And we can do that, right? Um, a- and add more layers, but that's the one thing I don't like about the Mapbox backgrounds, is that you can't do anything with them. Like, you can't play with them unless you go to Mapbox to customize it and then bring it over, right? But part of my own kind of design process is I like to do it kind of together, right?
25:35 Like, I would like to adjust, be able to adjust the background as I'm playing with the data itself, right? And so that's just kinda my own process, right? But you can go to Mapbox, create a custom map background, bring it into Tableau, um, that's maybe more consistent with your, with your organizational brand, things like that. But, um, you can't do that kind of at the same time, right?
26:02 Okay.
26:03 All right.
26:04 So one thing that I can do, and I can… You know, you'll notice that there's a province, also geographic field as well here, but I can't use this necessarily to outline South Africa because not all the provinces are represented here in the data. So what I'll do is I'll bring in another data set. I'll bring in the same data set,
26:28 the provinces this time, and it's just a list of the provinces, right?
26:34 And using map layers, I don't know why Tableau does, has this little space thing here.
26:40 But, uh, adding, I'll add the province as another map layer.
26:48 Move it to the back.
26:54 Oh yeah, of course, again, not in Canada.
27:03 Okay. And then now we have the outline here. Now, now I can pl- kind of play with this a little bit more, right? Like, I can make this maybe dark. Let's, let's go dark with a kind of a, a light border, and then I can make the text
27:23 white.
27:26 Like so.
27:27 And then maybe the points white too, to kinda match.
27:34 Okay.
27:34 No, that doesn't work 'cause it, 'cause these, these labels are over here.
27:39 Um, let's stick with…
27:49 So you're seeing, you're seeing kind of maybe a glimpse at my process. Like, I'm just gonna, I'll probably play with the labels for another hour. So if you, if you're, you're ha- you know, you can watch me do that or I'm gonna move on, okay? So let me d- let me just go back to kind of the original. Um, okay.
28:08 I, uh, l- we'll stick that for now and then let me, uh, uh, maybe make this a little lighter and this a little darker. Border a little darker.
28:18 Yeah. Okay. So let, let, let's stick with that.
28:21 Maybe even lighter. But, but now, what I'm trying to do now is, you know, I've added another sense of context, right? Uh, I have kind of Africa in the background, but specifically my story is about cities in South Africa, right? So now I've included South Africa and really highlighted that in the map, and then now we see the cities on top of that as well.
28:45 So, so now we have kind of these fi- three layers, right? Provinces, cities as graduated circles, and then cities as points on top of another fourth layer, which is the Mapbox background layer.
29:01 Okay?
29:02 So far, so good?
29:04 Right. And so this is just my maybe one way, you know, and obviously there's no right way, right? This is just one way that, you know, in my mind, I'm focusing on South Africa, so I really wanna kind of bring out South Africa, and so that's how I did it, right? Kind of as a separate data set.
29:21 Um, and then with the cities on top and things like that, okay?
29:27 All right.
29:28 So let's quickly move on to the next map. So I, I already have the cities, so let me bring in the cities again.
29:35 Or actually, I could duplicate the… Let me duplicate the sheet 'cause we can turn off things.
29:42 Okay, so I already have a map there. What I'm gonna do is I'm gonna start turning off a couple things. Let me turn off province.
29:49 Uh, let's keep the province. No, I'll turn it off. And then I'm gonna also turn off the background. I don't like it.
30:00 And I'm gonna start bringing in my own base map, right? And so I was trained, when I, when I first started GIS, this was ArcGIS Maps.
30:09 Um, that was the software. And I, there was… I don't remember any such thing as base maps And so when we created a map, we actually created it from scratch. We brought in all the files, all the data that we wanted to create this, and including raster data, including satellite images if you wanted to include that, right? And none of that was online at the time, um, or available, um, through the software.
30:35 And so we often created things from scratch, right? So, um, so I'm gonna tr- you know, kinda emulate that a little bit here, and this is kind of m- been, been my process.
30:46 So I have a data set of African s- uh, countries, and so I'm going to bring that into Tableau, and you'll see that here.
30:58 Okay. And go to sheet one. And then I can… I'm just gonna bring that into
31:05 the field.
31:07 And this is, uh… And I'm gonna start labeling these because this is gonna get com- start getting confusing.
31:12 And I'm gonna bring this down to the bottom as well.
31:19 Okay.
31:20 And I'm gonna just cre- make it kind of dark or make, make, make it light, so at least we have that background. And then no,
31:30 no border. Okay. And I'm gonna use that same data set 'cause I wanna specifically pull out South Africa, right, as a country.
31:43 Um, and so this is gonna be South Africa.
31:48 And what you notice when you bring in the shapefile, you have this geometry field that you would be… That this is the field that you'll bring into the data, the, the vi- uh, the data view or the view, um, when you, when you're adding layers. Okay?
32:02 So if…
32:05 I think this is the field. If the name is South Africa, then geometry. Okay.
32:15 End. Okay, and so what I'm doing here is I'm just gonna create a geometry field that for just, uh, South Africa. Okay.
32:27 And you'll notice that it creates another geometry field, and I'm just gonna pull that in here,
32:34 and I'm gonna bring that down to right above South Africa. Now, once I've done that, now I can start using some spatial, spatial, um, spatial functions or spatial calculations. And the one thing that I'm gonna do, 'cause I wanna be able to adjust the size of this border right now, okay? Could just kind of play, be able to play with it, right? And so what I'm gonna do is I'm gonna use a function.
33:00 I'm gonna call this South Africa border.
33:04 And I'm gonna use a function called outline.
33:07 Okay.
33:09 And let me make this a little bit bigger too.
33:12 And all I'm gonna do is I'm gonna bring in that geometry field of South Africa, and what this is gonna do,
33:21 this outline, it basically returns a polygon, um, uh, it c- it com- it, it… Sorry. It converts a polygon j- and, and converts it to a line, which is, it basically the outline of, of that polygon or polygons, right? And so right now this is a single polygon. I'm gonna ret- it's gonna return the boundary, basically the border, uh, of South Africa.
33:48 Okay. I'm gonna push okay.
33:50 And then I'm gonna bring that into the field.
33:54 And you, you can't see it now, but I'm gonna bring it down here. But on- once I turn it dark, now you can start seeing it, right? And because it's a line, I can adjust the size of the line.
34:06 Okay?
34:08 All right. And so let, let's make it like this.
34:13 Like so. Okay.
34:15 And so this, uh, outline is an example of a spatial function that you can only do with spatial data. All right? And so not the geographic roles that we've been using up to this point. And so in this case, I, I brought in a, a spatial, a shapefile of the African countries, and then I'm applying… I, I'm able to apply the outline function or calculation to the geometry field.
34:41 All right?
34:42 Okay.
34:44 Next thing I'm gonna do is I wanna, I, I want to also outline the provinces, right? So I have a, a separate, um, set of shapefiles, um, and this is like kind of almost, uh, like, you know, the different administrative boundaries, and I believe I want number one.
35:04 Um, ad- administrative boundary one, which is the provincial boundaries or provincial polygons. So I'm gonna bring that into
35:15 Tableau, and that's right there. And then I'm gonna bring in
35:23 the provincial boundaries. And you can see those there, I believe, when we add these. These are provinces.
35:30 And that's border. Okay. And now I'm gonna bring that down below the cities as well.
35:34 And I'm just gonna kind of, uh, I'm gonna make this, um…
35:40 Turn the opacity to zero, and then I'm gonna make the boundaries
35:47 maybe a little light like that. Okay.
35:49 And then I'm gonna make the boundary for South Africa a little thicker.
35:55 And just from a kind of visual standpoint, I think this looks, I think this looks pretty cool, right? Like kind of the differentiation of the borders and the size of the borders, I think it makes a big difference, right? Clearly, the darker border, the thicker border around the entire boundary is kind of a bigger geographic unit, and then the smaller ones are kind of outlined by lighter lines, right?
36:23 Um, so let me make South Africa Make it
36:32 so.
36:34 And then I can also, you know, I can pr- uh, let me switch this to this, and I think hopefully y- you know, you're starting to get a gist of maybe a cleaner map because now I'm not relying on defaults, I'm not relying on the Mapbox. I'm really, as I'm designing, I'm kind of making these decisions on these are the things I really wanna draw out and pull out.
37:00 Um, and hopefully it results in a cleaner map, right, as a result.
37:05 Um,
37:10 okay.
37:11 So far so good.
37:13 How… S- Suzanne, are we doing okay with time?
37:17 Suzanne Edlmann: Yes. We have 15 minutes left, um, but I think you can continue because there's no Q&A yet.
37:23 Dennis Kao: Okay. There's no questions? Okay.
37:24 Suzanne Edlmann: No.
37:25 Dennis Kao: All right. 'Cause everything I'm saying is very clear right now, right? Makes so- total sense.
37:29 Suzanne Edlmann: Yes. Thank you.
37:31 Dennis Kao: Okay.
37:32 Let me know if there's also, like, e- even Q&A, like, if there's kind of, like, even if it's like, if it just moves us in a tangent, I'm, I'm totally okay with that as well. Okay.
37:43 Um, so I have a couple things in plan in mind that might maybe help with the map a little bit and provide a little bit more context. But, you know, for general purposes, for most purposes, I think this would be a good kind of map, right?
37:59 Um, one thing that I wanted to kind of show you also is what if I turned off
38:09 the other out, o- other kind of bordering countries, and just keyed in on this, uh, on, on South Africa? And I think this works, um, especially if somebody's, you know, if, if your audience is specif- you know, is familiar with South Africa or, you know, is based in South Africa. But let's create a, an extent map, which is basically would be a map of Africa with South Africa, uh, with South Africa kind of highlighted, right?
38:38 And I can, and I'm thinking that because there's some open space here, I might include that map right here.
38:44 Okay? So let's do that right now.
38:47 So I have all, I think I have all the data that I need here, right? So I have Africa, so let me add that to,
39:01 to the view. Okay.
39:03 I also have South Africa,
39:09 like so. And then let, let me add the border just in case. Let me see if I, I need the border. Okay.
39:15 All right. And so what I can do with this maybe is I can…
39:19 Uh, let's make this dark,
39:23 the outlines, um, and I'll make this the same color as our data.
39:31 No border.
39:34 Change the opacity a little bit.
39:38 And then let me see if I, what happens if I add the border like that.
39:44 Okay.
39:45 And then what I'm gonna do is I'm gonna ch- turn off everything, um, the, in terms of the, um, background, so they will show just in case.
39:58 Um, get rid of all the lines.
40:03 All right.
40:04 Okay. Let's, let's go to the dashboard.
40:08 Okay. And then let's move that, and I'm gonna float it, um, right about here.
40:17 Oops.
40:26 Okay. And I don't like the sizing, so let me resize that.
40:34 Okay.
40:35 Now, what I'm seeing now is, like, I don't, I don't know if I need these lines. I'm not sure if I need this border.
40:42 So let's turn them off. Okay. So South Africa, I don't need this border here.
40:47 Um, and then I don't need the, the borders for the, the countries as well.
40:56 Boom.
40:58 So I think that works, right? I think, you know, we've…
41:02 So what, so, so some of the p- kind of thinking around this, right?
41:07 The, the focus is on South Africa, cities in South Africa.
41:12 All the other stuff around it, the bordering countries, the oceans, um, the terrain, the satellite images, they don't matter, right? So I turned them all off, right? Even the bordering countries, right? But I wanted to still provide the reader with some context, and so I provided them with a little extent map here that allows them to see, oh, and we're keying in on this particular area here, right?
41:39 So that's kinda often one way to kind of be able to do that, right? I think the advantage of doing it this way is that now there's not a lot, any noise around the, the key, the key information, right? So, you know, uh, when you have the Mapbox, like for, you know, I, I'm based in Canada, so whenever I map Canada or I, I'm from the United States, whenever I map United States, you know, Mapbox will provide Canada, Mexico, the oceans and things like that through the Mapbox background.
42:13 But if you're focused on the United States, for example, you don't need all of that, and people are familiar enough with the kind of the shape of the United States that you don't need to know that it's south of Canada or north of Mexico, right?
42:28 Um, and if your, your, your audience is specifically South African, you don't need kind of the other context here.
42:37 Um, but then you can always provide a little bit of context using an extent map or something like that, okay?
42:43 Um, there maybe a couple things that I might try out, for example. So one thing that I might try out is, let me add…
42:58 I wanna- let me see what happens if I add the water to this, like basically polygons.
43:05 Um, so let me add this.
43:12 And so, you know, what you've noticed is I'm just adding these as separate data sets. I'm just dragging them in and they'll, they'll a- uh, Tableau will, it'll connect to them as separate data sources, and you can just simply add them as another map layer.
43:29 Let's see. Does that look nicer? Does that look nice? Let me, uh, make this a dark blue and then border.
43:42 Does that add anything?
43:45 Maybe.
43:49 Let me look at the big map.
43:52 Eh, I don't know.
43:54 Maybe not. Okay. Well, we'll just keep that. So, so the cool thing is you can add them, and you can always turn them off, right? Like, you don't, like, turn off and on. Like, y- it's not, it's not, like, that much added kind of inconvenience. Like, you don't like them, you take them out or you just hide them, okay? Uh, let me try add the, the roads and see if that adds a little more context as well.
44:20 And what I'm doing with shape, if people aren't familiar with shapefiles, like, if you could take a look at this. So a shapefile is made up of… And, and, and the reason why it's kind of antiquated, it's, it's, it's made up of, like, five to seven different little data files, um, each of them representing different types of information that feed into a single shapefile when you visualize it in Tableau or GI- GIS tool, right?
44:46 So what I'm doing specifically is I'm pulling in the file. They all have the same name, you'll notice, and they're all together.
44:54 Um, and that's how the, the software will read them together.
44:58 Um, what I'm doing essentially is pulling in the one that has the shapefile extension, SHP, and I'm just pulling that into the view, and it'll basically connect to it.
45:11 Oh, this one's huge. This is, these are roads, right? So this is gonna be huge.
45:16 Um, but, um, and then you can a- then it gets added as a separate data source here.
45:22 Okay, so let me add the roads.
45:27 Just last thing.
45:31 Just because I g- I got the road data, I wanna kind of play with it and see if it, it works.
45:38 Okay, so these are the roads.
45:40 Um, but what I'll do is I will take out a few of those, and I, I know the codes, um, automatically, so let me pull that in. I'm just gonna take the, the first few.
45:56 Okay. Like so.
45:58 Really make them as small as possible. Maybe take away one more.
46:05 These codes are basically, um, major highways, major streets, all the way down to trails and things like that. So I'm just gonna play with, um, adding a couple more roads just to see how it looks and things like that, and then make these a little bit, um, the opacity really…
46:26 Yeah, and let's see. How does that look?
46:29 Eh.
46:30 Maybe, maybe not.
46:33 Okay.
46:35 All right. I think I'm done.
46:38 I'm ready to ship this. No, I'm just kidding.
46:42 Suzanne Edlmann: Thanks, Dennis.
46:42 Dennis Kao: Thank you so much for your time.
46:44 Suzanne Edlmann: Yeah. Actually, I wanted to just, um, pitch in and say there's five minutes left, um, that we can maybe just use if there's any questions. But-
46:52 Dennis Kao: Yeah, of course
46:52 Suzanne Edlmann: … yeah.
46:53 Dennis Kao: Happy to entertain.
46:53 Suzanne Edlmann: Thank you so much.
46:55 Yeah, thank you so much for this. Um, it's been very insightful. I think, um, I don't know how many people on this call have, uh, made use of maps in Tableau, and I think this is a very good illustration of what is possible, and you can start very simple and come to very interesting and visually aesthetic, um, visualizations very quickly.
47:18 Um, so thank you so much for the, the tips around that. Um-
47:22 Dennis Kao: Yeah, I know. And then, and just to wrap up, I have some, like, you know, they're not major tips. I think, uh, hopefully I cover all these. Keep it simple, do not accept defaults, take advantage of map layers. You know, and, and as you know, you can see me focusing on these kind of very little details and things like that. And just a quick plug, um, data maps, uh, TUG, is the only TUG that's focused on a single chart type.
47:46 Uh, Tori and I are the co-leads. Tori's on the ph- uh, yeah, I believe he's still on.
47:51 Uh, and, um, we have our next meeting in August, and it's a good one, I think, a- as you can see.
47:57 Okay.
47:59 Suzanne Edlmann: Awesome. Thank you so much for that, Dennis. Um, yeah, if you have any interesting links to post, you're welcome to, to send in the chat.
48:08 Um-
48:09 Dennis Kao: Okay
48:09 Suzanne Edlmann: … but I think, yeah, this one, if, if you are interested in the Tableau user group, please go and follow it. There's no, like, um, prerequisite. You can just, um, um, join it and, um, see the upcoming events around it.
48:25 Um, I just wanna see if there's any questions. Um.
48:30 Dennis Kao: Sahir. I think I see Sahir has a question.
48:33 Suzanne Edlmann: Yes. I do see that one.
48:34 Dennis Kao: Um, based on my experience, Tableau and GIS, has Tableau, as Tableau improves its spatial capability, do you see the gap narrowing?
48:44 Um, I think so, um, when it comes to visualization.
48:50 I think Tableau is still kind of behind the curve in terms of dealing with spatial data, although you'll notice in our next Tableau user, uh, data plus maps Tableau user group, we're gonna have somebody from Tableau covering Tableau Prep spatial functions. And so just, I think just recently a bunch of ta- spatial calculations and things like that have been added to Tableau Prep, and so there might be a lot more you can do with Tableau Prep.
49:20 But I think, so my workflow, what I often do is I will manage and clean and prep the data in QGIS because it's, you know, I, I, you know, it's, I think it's just ahead of the curve. You know, it's a specialized software that, that deals with spatial data. It's just easier to use and things like that, um, in terms of dealing with the prep, and so I might kind of filter some data into QGIS.
49:49 I might cut some, clip some. I might create some spatial, you know, data as well in Q- G- GIS. And then I'll bring it into Tableau because I think Tableau is really good at visualizing with the visualization piece. And also when, when you pull, put it onto a website, embed it, once you publish it online, then you have the interactivity kind of added kind of functionality.
50:16 And so, um, Tableau's really good at visualizing, uh, maps and with the interactive stuff.
50:25 Um, but I will still probably prep all the data in a QGIS tool or a GIS tool like QGIS. Yeah.
50:33 Suzanne Edlmann: Thanks, Dennis.
50:34 Dennis Kao: Hope that helps.
50:38 Suzanne Edlmann: Any more questions before we end off?
50:42 Dennis Kao: Um, so Mark had a comment, having a great eye is difficult to teach.
50:46 Um, to be honest, I, I didn't know I had this whole kind of creative side until I started just started to create maps in Tableau about six years ago, and I just discovered this creative side that I, I thought I never had. And I think the eye, having a good eye, I think comes with pr- I think it comes with practice. Like you kind of develop your own style.
51:10 You adopt or you're inspired by other people's maps or visualizations. You borrow kind of some ideas, and then you kind of merge those with your own, and it eventually becomes your own design kind of style.
51:24 Um, I think, I think that, you know, that great eye, I think, you know, maybe there is some, some,
51:33 you know, natural, you know, being, you know, being a natural at it. There might be s- you know, something there, but I think it's also something you can practice and, and develop over time. And over this past six years, I think I've really, you know… I, you can see, I think my maps have improved over the course of the past six, six years, and you can see that both in my professional maps as well.
51:59 Suzanne Edlmann: Well, thanks so much, Dennis, for your time.
52:01 Dennis Kao: Yeah.
52:02 Suzanne Edlmann: And thank you everyone for joining the call. Um, it was very insightful, and we are excited to have you in our user group again in the future.
52:10 Dennis Kao: Yeah.
52:11 Suzanne Edlmann: So thank you for that.
52:11 Dennis Kao: Thank you, everybody.
52:13 Suzanne Edlmann: All right. Thanks everyone. Bye-bye. Thank you.
Shorts for this TUG
- Why Tableau is my favourite tool for maps
- Map layers made Tableau feel like GIS
- Mapping a place you don't know
- Free spatial data is everywhere
- A quick map from geographic roles
- The trouble with Mapbox backgrounds
- Build your own base map
- Border weights show the hierarchy
- Drop the context, add an extent map
- What a shapefile really is
- Is Tableau catching up with GIS?
- A good eye comes with practice