How to Connect to Different Data Sources in Tableau: A beginner's guide

0:00Hey, it's Tim here. Connecting to data is

0:01one of the most basic things in Tableau.

0:03And in this video, I've taken a snippet out

0:06of my Tableau desktop crash course

0:08to essentially give everyone a great

0:10starting point for understanding how Table

0:12au connects to pretty much any data source.

0:14We cover CSV's, Excel files, and we even

0:16look at a database as well as describing

0:18some of the other sort of file types that

0:20you can connect to.

0:21So as ever, let's get stuck in.

0:23So here this time I'm on my Mac, just so

0:26everyone's benefit.

0:27I'm on my Mac and everything looks exactly

0:30the same.

0:31Maybe sort of Mac specific things are on

0:32this particular page.

0:34And you can see that I have this connection

0:36pane on the left hand side.

0:38Okay. And this connection pane is pretty

0:40important because depending on what you're

0:42trying to connect to, you need to go to the

0:43right section to select that file.

0:45For argument's sake, let's say that I want

0:48to connect to a CSV file, like a text file

0:51with some information in it.

0:52I can go ahead here and select text and

0:54what it will do is it'll open up my

0:56connection window, my file window, sorry.

0:59And you'll see here that at the bottom it's

1:01looking for these types of files.

1:03So if you're not sure what type of file

1:04Tableau is expecting to find when you

1:06select an option, you can just go down here

1:08to the bottom.

1:09And on Windows it's pretty much the same,

1:10but it just looks like a Windows interface.

1:12It will tell you what files it's looking

1:14for.

1:14And that's why these are all grayed out

1:16because none of these files match the

1:18criteria of this list.

1:20And that's essentially what you need to

1:21sort of look out for.

1:23Now, how do you know the file extensions

1:24for these things?

1:25Well, it's just a little bit of experience.

1:28Like Excel is .xlsx or .xls or whatever

1:31Excel has been over the years.

1:34That's what it's going to be.

1:35That's a very well-known file format.

1:37Text files on the other hand are a little

1:40bit broader.

1:41You can have .txt, .csv, .tab, .tsv.

1:44TSV is tab separated values.

1:46CSV is comma separated values.

1:49And it goes on and on.

1:51Okay.

1:53The JSON files.

1:54Now this is a web optimized format.

1:56So you can see here it's only looking for

1:57one specific file.

1:59I'm looking right here.

1:59If you're wondering where I'm looking, I'm

2:01just looking here where it says JSON file.

2:03And that's one specific format.

2:05That's known as JavaScript Object Notation.

2:08I think that's what...

2:09No, JavaScript.

2:10Yeah, it is JavaScript Object Notation.

2:12And what that means is essentially it's

2:14data stored in a specific way, typically

2:17from web-based systems.

2:19So if you go and export something from an

2:21internet-based system and you just ask for

2:23the raw data, typically it will give it to

2:25you in JSON.

2:26So you can connect directly to that as well

2:27.

2:27I hit cancel.

2:29You can connect to PDFs.

2:31Now this is a bit hit and miss.

2:32The PDF scraping, you know, it's got a

2:34capability built into Tableau that uses a

2:37scraper that goes and looks into the PDF

2:40and tries to find the data across multiple

2:42pages.

2:42But it can be hit and miss.

2:44It's not always reliable.

2:45So try it.

2:45If your PDF has data you think Tableau can

2:47grab it, try it first.

2:48You might save yourself a ton of time

2:50copying and pasting.

2:51But if it doesn't work, then yeah, you

2:52might have to resort to something a little

2:54bit more advanced or simpler.

2:55Just copy and paste yourself.

2:56Spatial files are quite an interesting

2:59addition to Tableau.

3:01They've been in Tableau forever.

3:03I don't know why I say addition, but they

3:05can be connected in sort of a couple of

3:07ways.

3:08Now spatial files are typically this whole

3:11range of files.

3:12I won't go through each and every one of

3:13them explaining them.

3:15The ones you're typically going to see used

3:18commonly with Tableau are these Esri shape

3:21files dot shp, GR JSON files, which can

3:23come from web applications like Google Maps

3:25and so on and so forth.

3:27And then you've got KML files and a couple

3:30of other files that sometimes come out of

3:33similar systems.

3:35So KML, KMZ.

3:36You might have this MIF and dot tab as well

3:39.

3:40Those are very specific.

3:41And then zip files typically also contains

3:44spatial data within them as a package.

3:46And that's why it comes as a zip file.

3:48So Tableau is able to sort of work with all

3:49of that.

3:50That's why here you can actually see that

3:52it's highlighting the files it can see.

3:54So all this time I've been going through

3:55the files.

3:56It's not been capable of connecting to them

3:58here.

3:59You can see that it's happy to connect to

4:00these two files.

4:01So let's go ahead and find some data to

4:04connect to.

4:05Let's go ahead and to do this.

4:07I'm just going to go to a website and try

4:09and find some data.

4:10I wanted to give you sort of an honest

4:11experience of connecting to data just found

4:14on some sort of website.

4:15So I'll go to Kaggle.

4:16Kaggle is really, really good.

4:18It's gone to a dataset that I used a long

4:20time ago.

4:21So let's go back and let's go to the Datas

4:24ets tab and we'll try and see if we can

4:26filter this.

4:27And you can filter by specific file types.

4:30Let's just select the CSV and hit apply.

4:32And you get sort of a list of different

4:35files.

4:35I'll connect to this top 100 Spotify songs.

4:38This looks pretty good.

4:40And you can get a little preview of the

4:42data.

4:42So let's go ahead and download this.

4:44It looks pretty small, but let's go ahead

4:47and download it anyway.

4:48Oh, I need to register.

4:49All right. Let's go ahead and sign in with

4:51Google.

4:51And once we've done that, I should be able

4:55to download this.

4:56Let's go ahead and do that.

4:58And yeah, I'll save it to my desktop just

5:00to keep things nice and easy.

5:02Once it's on my desktop, I'll show it in

5:05the finder and I will unzip it.

5:10OK. So now that it's been unzipped, where

5:12is it gone?

5:13Let's see. Where is my.

5:15Where has it put the file?

5:18It's not. It's not. Doesn't seem to be

5:23showing up.

5:24Let's click out and click back in.

5:25Right. Let's try this.

5:27Let's just I'm pretty sure it's on my

5:29desktop.

5:29Let's just go try and find it.

5:30The CSV files. If I select text file and I

5:33go to my desktop, you can see here.

5:37There it is, a list of most stream songs on

5:39Spotify dot CSV.

5:41And if I just click out of it, you can see

5:43all these other files don't match that

5:45criteria.

5:45That file does.

5:46I select open and you go into what I call

5:49the connection interface.

5:51Now, let me just give you a quick overview

5:53of this interface.

5:54The connection interface is broken into

5:56what I call three sections.

5:57You've got the preview section on the

5:59bottom.

5:59The reason I start with the previous,

6:00because people always want to see something

6:02familiar first.

6:02So this is where to go to see your data

6:04that you've just connected to.

6:06You get a list of the field names on the

6:08left hand side of that preview pan so you

6:10can see what that is as well.

6:12And then on the top, you have what is

6:14called I'm going to call it the data

6:16modeling windows.

6:18Oh, sorry. I didn't like that.

6:20So let me take my watch off so it doesn't

6:21try and interfere with the live stream.

6:24This is called the data modeling window.

6:27And in this window, you can essentially

6:30build data models and work with your data.

6:33I'm trying so hard not to get into really

6:35advanced topics.

6:37So bear with me.

6:38But I'll explain what a data model is very,

6:41very briefly in a second.

6:42You don't need to worry about it.

6:43And on the left, you have what is

6:45essentially a list of your connections and

6:47the resources that you can bring into your

6:49connection and use.

6:50OK, so what is the workflow for connecting

6:53to data?

6:53Well, as soon as you connect to data, the

6:56first thing you should look for is what is

6:58the connection called?

6:59This is essentially going to match the name

7:00of your file.

7:01And if you want to, you can change this.

7:03The easiest way to change this is just to

7:05double click here at the top and you can

7:07change the name.

7:08I call this Spotify.

7:09Spotify.

7:11Hit enter.

7:12And you'll see that that changes.

7:13But it's not going to change the actual

7:16name of the file that I'm connecting to.

7:19So the name of the connection is changed up

7:21here, but it's not going to change the name

7:22of the file that I'm actually connected to.

7:24That is always going to be the name of the

7:26file.

7:27And once I've connected to the file just

7:29below it, I'll highlight this in red.

7:32You have essentially the different things

7:34that can be in that file.

7:36So when we connect to an Excel file soon,

7:38you'll see that this is slightly different.

7:40But here I've just got this one file here

7:43that you can see a list of most listened to

7:46songs on Spotify, which is essentially the

7:49file, the CSV file.

7:51And then on the bottom, I'll highlight this

7:53in green, you have essentially a couple of

7:55new features.

7:57So you have the union, which allows you to

7:59bring dates together.

8:00We won't cover that today.

8:01That is, again, it's not really an advanced

8:03topic, but it's just not sort of pertinent

8:05in this particular sort of crash course.

8:08And then the new table extension, which

8:10really is advanced.

8:11Don't need to don't need to worry about

8:13that at all.

8:14But very briefly, this allows you to use

8:17analytical applications to bring in data

8:19from other systems like Python and R to

8:21bring data alongside your normal data.

8:24OK, so that's the left hand side of the

8:26connection window.

8:27Now, when you've done that, let me go ahead

8:29and remove this.

8:30When you've connected to your window, you

8:32see when Tableau automatically does

8:33something for you when you connect the

8:35first time, it goes ahead and bring

8:37something into the space for you.

8:39And as soon as you drop it down, you get

8:41the preview we saw before.

8:43So in essence, when you connect to

8:44something for the first time, it's already

8:46done for you.

8:47And so you're probably thinking, well, what

8:48do I need to do?

8:49Well, it's already been done for you.

8:50It's right here.

8:51But if you don't see that or if you want to

8:54bring in a second item, what you need to do

8:56is just go ahead, grab the file and bring

8:58it in and drop it in.

9:00And now you get a preview.

9:01The other thing is you might want to see

9:03your data in a sort of more spacious window

9:06.

9:06And if I hover over the farm, you can see

9:09there's a little tiny box that appears just

9:10here where I'm highlighting.

9:12It's disappeared because I'm not hovering

9:14over it.

9:14But now if I hover over it, you can see it

9:16's right there, this little tiny table icon.

9:18And when you hover over it for long enough,

9:20it says View Data.

9:22You can open it up and go ahead and view

9:24the data.

9:24So if I just go ahead and drag this out, it

9:26's not letting me drag it out.

9:28There you go.

9:29Let me drag it down.

9:30No, it's not going to let me drag it down.

9:32Just going to let me drag it out.

9:33That's absolutely fine.

9:34I won't complain.

9:35And it shows me the data.

9:36So essentially you've got one, two, three,

9:39four, five columns.

9:40OK.

9:41Now that I'm here, it's important to notice

9:43a couple of things that Tableau has done.

9:46Firstly, it's gone ahead and looked at our

9:48data and understood that each column

9:51represents different types of data.

9:53So types of data just mean numbers, text,

9:56values.

9:57And you can see that it's calling this one

9:59ABC, this one ABC, and this one ABC.

10:03Essentially, it thinks that these three

10:05columns contain text.

10:07OK.

10:08Now we can see that this column here

10:10contains dates.

10:11So we're going to need to correct that.

10:13We'll come back to that in a second.

10:14The other thing, though, is that this

10:16column here called rank, it's seeing

10:19contains numbers.

10:20And you can see it's got a little hashtag

10:21for numerical values.

10:23And the same again for streams in billions.

10:26So although this number says two thousand

10:28and fifty nine, I'm not sure if this two

10:31thousand and fifty nine.

10:33I don't know if this is two billion and

10:35fifty nine or if this is two thousand and

10:37fifty nine billion.

10:39I hate this kind of stuff.

10:40I think this is two billion streams.

10:42We can go on Spotify and have a look.

10:43But I'll just assume this is two point

10:45something billion streams.

10:46OK.

10:46So all of that information is called

10:49metadata.

10:50Metadata is essentially information about

10:53data.

10:53It's essentially the core information that

10:56tells you what is in your data.

10:57And you can then use that metadata to make

10:59sure things behave the way they should.

11:01For example, numbers should be treated with

11:03numerical values.

11:04Text should be treated with text behavior

11:07and text values in essence.

11:09OK, so how do we change this?

11:11Well, if I close this preview window, you

11:14'll see that just over here on the bottom.

11:16Let me just bring this pane to the left.

11:18I have the same preview that I've just seen

11:20, but it only shows me the few hundred

11:23hundred rows.

11:24I know that because it says one hundred

11:25rows right here.

11:27So if I set that to a thousand and enter,

11:29it will show me a thousand rows.

11:31It's still showing me one hundred rows.

11:32I don't know why it might be above on my

11:34Mac version, but it should show you

11:36whatever number of rows you type into that

11:38space.

11:38Now, when I go to this column here at the

11:40top, you can see that I can actually click

11:42on these as well.

11:43So let me go to the date column here.

11:46Release date.

11:47I'm just going to tell Tableau, look, can

11:49you change this to another data type?

11:51And when I click on that, I get all the

11:52data types that are available.

11:54So I get number, whole numbers, date, time,

11:57date, string, spatial, boolean.

11:59These are types of data.

12:01And if you're not sure what these are, go

12:03ahead and just Google them.

12:05They're pretty straightforward.

12:06Literally, once you know them, you know

12:07them.

12:07You don't need to learn them again.

12:08But they're pretty straightforward.

12:11In this particular case, I know that this

12:13column that I can just see right here when

12:15it says 29th of November, that should be

12:17just a date.

12:18There's no time.

12:19There's no like 29th of November, midday.

12:22It's just a date.

12:23So if I go ahead and click date, Tableau

12:25does something very smart.

12:26It goes and processes all that text and

12:29tries to make sense of the date.

12:31And because Tableau is good, it's actually

12:33gotten very good at that, that I'm

12:35confident that it's got it right.

12:36So you see here it says 29th of November,

12:382019.

12:39I'm in the UK.

12:41So this is the correct way of doing dates.

12:43Don't come at me if this looks completely

12:46wrong to you.

12:47This is the correct way of doing dates as

12:48far as I'm concerned.

12:50And so now it's changed that to dates.

12:53You can see that we have a calendar item

12:55right there ready to go.

12:57And this has been understood and everything

12:59's going to work nicely.

13:00So we can go ahead and start using this in

13:02our visualization.

13:03Now, if I wanted to bring another data set

13:06in, I could.

13:07But again, this is a crash course.

13:09We're just going to work with single data

13:10sources for now.

13:11I've done many videos and other people have

13:13done many videos about how to bring two

13:15data sources together.

13:17When you're doing that, you're typically

13:18doing something called a union or join.

13:20And if you work with Tableau, there's also

13:22this new concept called the data model.

13:24I've also done a video on that.

13:26So go and research that.

13:27But for now, assume you're connecting to

13:29one data source.

13:30You've got one connection.

13:32Everything is set up nicely.

13:33Now we're ready to visualize our data.

13:36What we can do is we can just go down here

13:38and Tableau is kind of telling you this the

13:40whole time.

13:41Go to this step to go to the worksheet and

13:43start visualizing something.

13:46So when I go ahead and click on the space

13:48where it says sheet one, we're into the

13:50building window.

13:51We can actually start building a data

13:53visualization.

13:54And now I'll go through this in a second.

13:57But I want to show you a couple of other

13:58ways of connecting to data as well.

13:59So that's pretty much the flow for

14:01connecting to data.

14:02Now, let's say you want to go back to the

14:04data connection window.

14:05You maybe realize that something is wrong

14:07and you need to change the file.

14:09Well, you can do that at any point.

14:11Just because you started building the

14:12visualization doesn't mean you can't change

14:15what you've already connected to.

14:16So to do that, you just want to go back to

14:18this data source tab here.

14:20So if I go ahead and select that, it takes

14:22us back and we're pretty much golden.

14:24We can just start changing everything again

14:26.

14:27We can rename it, work with it, but

14:28everything's going to work as you'd expect.

14:30It's going to be pretty straightforward.

14:32So that's essentially the first example of

14:34connecting to a file.

14:36We're connected to a text file.

14:37We're ready to work with it.

14:38We're ready to visualize it.

14:39We can kind of put a nib on that, bow it

14:41off and say we know how to connect to text

14:43files.

14:44Okay, the next file I want to show you is

14:47an Excel file.

14:48And Excel files are the most common files

14:50you connect to in Tableau.

14:51I hate to say it.

14:52People should be using databases, but Excel

14:54is just so common.

14:55So this is what you're going to have to

14:56connect to.

14:57So how do we connect to that?

14:59Well, I'm going to show you how to do that.

15:01Having connected to one connection already.

15:03You see, if I go back here and I start

15:05trying to add a connection here, Tableau is

15:07going to get confused because it might

15:09think that we want to bring the data

15:11together.

15:11We want to kind of put these two files

15:13together.

15:14I don't want that at all.

15:15I want a separate file in my workbook to

15:17work on a separate visualization completely

15:20.

15:20So how do I do that?

15:22Well, if I go back to Sheet1, let's say I

15:24've built a visualization.

15:25You can see that here at the top, I have a

15:28plus icon next to a cylinder.

15:30Now, a cylinder is typically the icon they

15:32use for databases pretty much everywhere.

15:34But go ahead and click on that.

15:36You'll see that I get the same window that

15:38we got when we were connecting to data at

15:41the start.

15:42So that connection window also lives here.

15:45Essentially, it should be very familiar

15:46with that.

15:47And this time I'm going to choose the

15:49Microsoft Excel file.

15:50OK.

15:51When we choose that, I want to go over to

15:53my documents folder.

15:54And I want to make you aware of a folder

15:56that Tableau installs on your machine in

15:59pretty much every case, unless you work in

16:01an enterprise organization where they've

16:03decided to take that file away from you.

16:06And it's called my Tableau repository.

16:09This folder right here.

16:10Now, this repository file comes with every

16:12installation of Tableau.

16:13And if you double click it, you'll see that

16:15it actually has a few things.

16:16It has some sample workbooks, has logs.

16:19If you ever have an issue with Tableau,

16:21this is where it will drop them, has data

16:22sources that you might have saved and has a

16:24bunch of other things.

16:26OK.

16:26One of the things I like about this is in

16:28that folder, you should have another folder

16:31with a version of Tableau you have

16:33installed.

16:34Now, I only have 22.4 on this particular

16:37setup.

16:37So I'll go ahead and double click that and

16:39then I'll go ahead into that file and I'll

16:41look at the English version of the data

16:43sources that I have.

16:44And you should also find in your version of

16:47this folder in your documents folder this

16:50file here called EU Superstore XLS.

16:53Now, this is a great file because it allows

16:55me to show you something.

16:57And I know for a fact that you should also

16:59have this file.

17:00If you don't have this file, I'll put a

17:02link to this file in the documents so you

17:04can find it.

17:05The thing to note, though, is that

17:07depending on your version of Tableau, these

17:10files are different.

17:11So if we start visualizing something and

17:13you see a slightly different number, don't

17:16worry.

17:16We've all got the same data source.

17:18It's just that they change them from

17:19version to version.

17:20They may be tweak a number here, tweak a

17:22number there.

17:23Sometimes the titles don't quite add up.

17:24But generally speaking, all the columns are

17:27the same.

17:27So just follow along with columns.

17:29And as long as it looks the same, you're in

17:31good company.

17:32And we've got two versions of the file.

17:34We've got a European version and American

17:36version because most of the YouTube

17:37audience is American.

17:38I'll go ahead with American audience.

17:40And the second biggest country on my

17:42YouTube channel is India.

17:43There is no Indian version of this.

17:45Maybe we should make one as a nice little

17:46side project.

17:47Let's make an Indian Superstore version

17:49with Indian locations that are more

17:51specific to the Indian market.

17:52Maybe we should do that.

17:53Let's go ahead and select the sample files

17:56like open.

17:57And when we do that, Tableau thinks about

17:59it and it takes us back to the connection

18:01window.

18:02So you're probably thinking, well, hold on.

18:04Last time we had the CSV file.

18:07Where has that gone?

18:08Well, it's not disappeared.

18:10What it's done is it's created a new

18:12connection.

18:12And the new connection allows you to

18:15basically switch between this connection

18:17and the previous one we made with the CSV

18:19file.

18:19And if you just go to this top little drop

18:21down, you can see that the one we renamed.

18:23I just go here.

18:25You see the one we renamed previously

18:26called Spotify is right there.

18:28And if we click on it, we go back to it.

18:30And if we click on this cylinder again and

18:32go to the new one we've just connected to,

18:33which is an Excel file, we go back to that.

18:36So you can switch between all these

18:37different connections just right here.

18:39OK.

18:40Now, if I go back to the Excel file, you'll

18:43see that it says right here that it's an

18:45Excel file.

18:45Let me change to my red annotate.

18:47I prefer that.

18:47Now, we've selected this Excel file and in

18:51the Excel file it has three tabs.

18:53If you're familiar with Excel, we have tabs

18:55across the bottom.

18:56You have, well, you just have tabs. Depends

18:58on what they are, but each of them could

19:00contain information.

19:01And so what Tableau is telling me is that

19:03this file contains three pieces of

19:05information.

19:06It has an orders table, a people's table

19:08and a returns table.

19:10And if I'm not sure what they look like, I

19:12can of course remember I can go ahead and

19:13preview them just by clicking on those and

19:16I can see this data actually contains

19:18information about orders made at my Pacific

19:20store.

19:21Let's go ahead to the people table.

19:22Connect to that.

19:24Also good.

19:26Returns.

19:26Connect to that.

19:27Also good.

19:28So this just allows us to look at the data

19:30sources and understand what's going on.

19:32OK.

19:33Now, you'll notice that as I connected to

19:36each of those, these icons changed.

19:39I don't know if anyone noticed.

19:40If you don't believe me, rewind and look at

19:41the icons again.

19:42These icons have changed and they've gone

19:45to a green square with a little sort of tag

19:48.

19:48And the tag lets you know that this data is

19:51actually coming from a named range inside

19:53of that Excel file.

19:55If you're not familiar with name ranges, I

19:56've done a whole video on this.

19:58So go ahead and look at the video where I

19:59talk about name ranges and Tableau.

20:01I talk about this in a lot of depth.

20:03But the key thing here is that when you're

20:05working in Excel, in fact, let me just go

20:07to Excel and I'll show you.

20:09If I go to Excel here, I have it open.

20:10If I say make a table, let's make it.

20:14Oh, my cursor is not working.

20:15There we go.

20:15Let's make a table of fruits.

20:18I've got my annotator on as well.

20:20So let's go ahead and say Apple and Pear.

20:25OK.

20:25And I can say this costs £10, very

20:27expensive Apple and this costs £5.

20:30OK.

20:30Now that's a piece of information.

20:32That's data.

20:33Now, if I highlight these, OK, and create

20:36what's called a name range, if I just go to

20:38the data tab just up here.

20:40And I can never remember how to do this.

20:42Where is the name range options?

20:43The actual easy way to do this is just go

20:47to the home and format it as a table, which

20:51is where is the format as a table option.

20:54My God, I'm really bad here.

20:57Well, I found the name range options

20:58instead.

20:59I can't get this dark mode version of Excel

21:01.

21:02It's just clearly throwing me off.

21:04So if I select the table, selectify name,

21:07you can see that it tells me the name range

21:09here.

21:09It's just pointing to that sort of square.

21:11If I select OK, that becomes a named range.

21:14That is essentially what we have.

21:15You can see here the name range is called

21:18Apple and it starts with this row.

21:20And that is essentially going to be visible

21:21in Tableau.

21:22As Apple, when I connect to this Excel file

21:25and start working with it, that's what a

21:27named range is.

21:27And people use them in organizations all

21:30the time to sort of section out data where

21:32you have multiple things on an Excel page.

21:34That's essentially how to connect to each

21:36and every one of those individual tables.

21:38And if it doesn't exist, you can go ahead

21:39and connect one.

21:40It allows you to pick out data from a page

21:42full of information.

21:43OK, but for now, what I really want to look

21:47at is orders.

21:48If I just scroll up, you can see that we

21:50have the orders view over here.

21:52What actually happened with this is the

21:54icons didn't change.

21:55I scroll down. That's what must have

21:56happened.

21:57This interface is...

21:58What is happening with Tableau interfaces

22:00recently, honestly?

22:02So this never used to scroll down.

22:04This just used to move down to make space

22:05for it.

22:06But it's deciding to scroll.

22:07So it didn't change icons.

22:08Anyway, here's the orders table.

22:10To bring it in, we just drag it in.

22:12Like I said, showed you before, we get the

22:14same preview as we've had before.

22:16Our data is here. It's pretty good.

22:18We can customize any of these as well.

22:20We can do anything we need to do, and it's

22:23pretty much good.

22:24Now, the final thing I'll show you is that

22:26there are some data prep capabilities in

22:29this window.

22:29If I go to this little dropdown, you can

22:32see that I have the ability to split and do

22:34custom splits.

22:35Essentially break out the data and in some

22:37cases even pivot the data if I need to do

22:40that.

22:40So in this case, if I just select split,

22:42Tableau looks at that column of information

22:45and it automatically splits it out into

22:47three columns.

22:48You can see here at the very end, it's

22:49added them in.

22:51And a little clue to let you know that

22:52those have been created here is that if you

22:54look here at the very, very top,

22:56just in this section, you'll see that these

22:59have a blue marker and these don't have a

23:01marker.

23:02The blue refers to this blue over here.

23:07So when you're connecting to multiple

23:09things, they might have different colors

23:11and those colors will show up on the

23:13columns to let you know which table they've

23:16come from.

23:17That's basically it.

23:19So that's why you don't get a blue icon

23:21here at the top, because these have been

23:23created inside of Tableau rather than just

23:26in general.

23:26So now that we've done that, we're pretty

23:29much good to go.

23:30We can go ahead to sheet one as Tableau

23:32hints us to.

23:33So I'm just going down here, selecting

23:35sheet one and we're back here ready to

23:37build some data, ready to work with our

23:39visualization.

23:40And we're pretty much good to go.

23:42If I look here on the top left, I now have

23:44two data sources, Spotify and Superstore,

23:46and that's pretty much everything you need

23:48to know about that.

23:49OK.

23:50The very last thing I'll show you how to

23:52connect to is a database.

23:54So you've connected to a flat file, you've

23:56connected to Excel.

23:57The last thing I'll show you is a database.

23:59Databases are pretty common at work, so it

24:02would be a pretty bad tutorial to show you

24:04that.

24:05So let's go ahead and connect to a database

24:07that I have access to.

24:09Now, to do this, I need to be super careful

24:11that I don't share my credentials on screen

24:13.

24:13So let me just give me one second and I

24:15just don't give you all my information to

24:18my database and I make sure that I have it

24:21in another window and I can type it in.

24:24Appropriately as and when I need to.

24:28So let me just where is this?

24:30Where have I put this data?

24:33Honestly, I should be more organized.

24:35Here we go.

24:35So I think I think we're OK.

24:37One thing I'm slightly concerned about is

24:39that I might have security settings on my

24:41database to stop me connecting because I'm

24:43not where I'm supposed to be.

24:44Long story short, for security reasons,

24:47sometimes it's good to tell your database

24:49not to let you connect from certain

24:51locations because those aren't locations

24:53you're supposed to be connecting to the

24:54data.

24:54So for this, what I'm going to do, I'm

24:56going to move this.

24:58Can I move this window to another window?

24:59Yes.

25:00I'm not going to show you this until I

25:01filled it in because I don't want you guys

25:03getting all my database details.

25:04So let me fill it in off screen.

25:07Apologies.

25:08Probably the worst.

25:09This is the worst live streaming thing to

25:13do.

25:13But you can't dynamically fade out stuff.

25:17I trust you guys won't be stealing my email

25:20and everything, but I need to just be able

25:23to log into this database and do that.

25:26So all I've done is I've typed in the

25:27credentials.

25:27Now, I've actually done a video on how to

25:29connect to Snowflake.

25:30So you can go watch that video where I have

25:32blurred everything out.

25:33But as soon as I connect to a database, you

25:35'll see I get a different kind of window.

25:38It's not like what we've seen before.

25:39For Snowflake specifically, I get what's

25:41called a warehouse.

25:43And each database has its own sort of setup

25:45.

25:45So you might get different things and you

25:47might have like a warehouse, but warehouse

25:49means something else.

25:50You might have a data lake.

25:51You might have all of these terms.

25:53You might have schema as well as an option

25:55that turns up in here.

25:56Depending on the database, they're all

25:57slightly different.

25:58This follows the terminology of that

26:00database.

26:01So I'll go select my warehouse.

26:03In Snowflake, warehouse just means how much

26:05computing power you want to use.

26:06So I have one called compute.

26:08This is my own database.

26:09I use this to run this YouTube channel and

26:11everything that goes on with my channel.

26:14And then I've got a demo database, a Table

26:17au demo database with demo Tableau data.

26:19So if I go ahead and select that, that

26:21works.

26:21Then I can go ahead and select the schema.

26:23This is the public schema.

26:25And in the public schema, you can see that

26:27I have three pieces of information.

26:29I have my employees table, my invoices

26:31table and my orders table.

26:33This is all dummy data, so there's nothing

26:34sensitive here.

26:35My orders table, I can preview it just like

26:38I showed you before.

26:39And it looks exactly the same.

26:40So these options are just slightly

26:42different because I'm connecting to a

26:43different type of data source.

26:44But if I go ahead and close this, you'll

26:47see I also get an option to do custom SQL.

26:49Now, if you're the kind of person who just

26:51knows how to write their SQL to connect to

26:53a data source,

26:53you can actually go ahead and just bring

26:55that in and paste your SQL in here.

26:58And Tableau will go and run that query to

27:00return the data that comes back and use

27:02that in here.

27:03The advantage is you could do a lot of data

27:06preparation in this window.

27:08But a note about custom SQL, it can

27:10sometimes slow Tableau down, especially if

27:12you're using a live connection.

27:14So just use it with caution.

27:16And the other nice thing is with custom SQL

27:18, you can insert a parameter that's

27:20controlled by the user that actually runs

27:22in this window.

27:23So if you've got a live connection, you

27:25want to give the user the ability to choose

27:28some sort of variable from the database.

27:30So your database is not chucking everything

27:32out the user.

27:32You can actually use a parameter in this

27:34window to do that.

27:35That's just worth noting.

27:36But nonetheless, same as always, drag my

27:39orders table in.

27:40It looks exactly the same as we've been

27:42using before.

27:43Table there, preview at the bottom, ready

27:46to go.

27:47Go to Sheet 1, exactly as you'd expect,

27:50connected.

27:51OK, so we've got three connections.

27:53We've got our Snowflake connection, we've

27:55got our Excel connection, and we've got our

27:57CSV connection.

27:58All three of them are up here at the top.

28:00Now, you're probably wondering, well, great

28:02, let's start visualizing our data.

28:05Let's start working with this information.

28:07Well, the tricky thing is actually we're

28:09not done.

28:10And let me just sort of take a break to

28:12sort of check what I've covered here.

28:14So the connection interface is done.

28:16Let's strike that out. Finding data to use.

28:19I didn't really cover that, so I'll come

28:21back to that.

28:21Connecting to data, I kind of feel like I'm

28:24doing that now, so I'll cross it out ahead

28:26of time.

28:26And yeah, the next thing we're going to

28:28come to is the Tableau extracts.

28:31You see, the easiest way to explain this

28:33without going into too much detail is that

28:36Tableau extracts are an optimized format of

28:39data.

28:39Right now, everything we're connected to is

28:43connecting live.

28:44The way I know that is that each of these

28:46cylinders are just cylinders.

28:49So these are all live connections.

28:50What does that mean? That means every time

28:52I do something in Tableau, Tableau is

28:55actually going to that file and querying

28:57the information.

28:58Every time I do something with this

29:00particular connection, it's going to Snow

29:02flake, querying the data and coming back.

29:04Now, that's fine.

29:06But if you're trying to build a dashboard

29:08at work, that's actually not fine.

29:10Because what happens if the database goes

29:11down whilst you're building something?

29:13What happens if you just want to build

29:15something quickly and you don't worry about

29:16the latency and the networking issues in

29:18your organizations?

29:19You just need to get on with work.

29:21And so Tableau has another way of capturing

29:23this, which is essentially by taking a

29:25snapshot of the data.

29:27What it does is it goes off to the data

29:29source.

29:29It takes a snapshot, but it remembers where

29:32it got the data from.

29:33And when it takes that snapshot, it means

29:35that it saves it into a more optimized

29:37format that allows you to do a little bit

29:39more with it.

29:40And it also works considerably faster.

29:43It's also a lot smaller.

29:44So a good example, if I took like, as I

29:47said, 200 megabytes CSV file, like a text

29:49file, Tableau would compress that down to

29:52about 10 megabytes.

29:54Much, much more portable, much faster.

29:57It's going to be much faster than opening a

29:58text file and looking through and querying

30:00it for information.

30:01And the added benefit is that extracts also

30:03allow you to do certain things that you can

30:05't do with just normal connections.

30:08Again, it's a little bit beyond the nature

30:10of this crash course, but you can do a

30:11little bit more with it.

30:13So how do you take an extract?

30:15Well, there's a couple of ways.

30:17If I go back to any one of these, in this

30:19case, let's go back to my Excel file.

30:21I can right click on it and I can select

30:24edit data source.

30:25When I do that, it takes me back to this

30:27window.

30:28And I wanted to show you another way of

30:29coming back here just so that you know that

30:30you can do that.

30:31OK. Now, up here on the top right, you can

30:34see there's an option that says extract.

30:37I completely missed this before because I

30:39wanted to wait until I was here to kind of

30:41talk a bit about this.

30:42And you can see the default option is live.

30:44But what I'm saying is you should be using

30:47an extract.

30:48OK. And so let's have a look and see what

30:50happens when you switch over to the right

30:51hand side.

30:52You see, when you go to the right hand side

30:55, it says extract will include all data.

30:58It's not created the extract yet.

31:00It's just telling you that this is going to

31:01include all data if you were to take one.

31:03But here's the advantage.

31:05You can change what data comes in.

31:07You can change what data comes into your

31:09snapshot just by selecting edit and tablet

31:11gives you this window.

31:13That allows you to choose what data you'd

31:14like to bring in.

31:15Don't worry too much about the stuff here

31:17at the top.

31:18All you're paying attention to is the

31:20filters pane here in the bottom.

31:21OK. When you select add, it shows you all

31:24the columns in your database.

31:25Let's say I only wants to bring in cells

31:27from a Pacific subcategory.

31:29I could go ahead and select.

31:30OK. Choose that subcategory.

31:33Select. OK.

31:34And now I've limited my data to just the

31:36art data for my extract.

31:38I've not deleted it from the database.

31:40I've not deleted it from the file.

31:42I've just brought in a tiny sample of that

31:44data because maybe this is all I'm

31:46analyzing.

31:46If I select OK.

31:48That's all fine.

31:49And now when I go back to my visualization

31:52to start using it.

31:53Tablet actually asks you hey where would

31:56you like to save this extract.

31:58And this is kind of confusing because you

31:59probably think you're like whoa whoa whoa.

32:01I thought you were going to save it in the

32:02workbook in this particular file.

32:04And you're going to take a copy of my data.

32:06But in actual fact, Tablo does need to

32:08write the file somewhere.

32:09Likes is going to write the file.

32:11So I always say to people look save it to

32:13your desktop because we're going to delete

32:14this later.

32:15I'll show you why.

32:16Save it to your desktop.

32:17And when it's done you know you have an

32:20extract.

32:21When you go to the top left hand side here

32:23and it has two cylinders with an arrow

32:25going from the first one to the second one.

32:27Essentially it's telling you that it's

32:29taken an extract and now it's using that

32:31extract.

32:31OK.

32:32And so if I go and ask Tablo hey what subc

32:35ategories do I have in my data.

32:37I'll just go ahead and drag subcategory

32:39onto text.

32:40You'll see the only data I have in that

32:42file is art.

32:43But if I want to connect back to the live

32:46data let's say the database is back up and

32:48everything is great.

32:49I can go back in right click on sick the

32:52use extract option and all the data comes

32:55back.

32:56OK.

32:56So just by switching that on and off I can

32:58switch between whether I'm using my

33:00snapshot or whether I'm not using my

33:02snapshot.

33:03OK.

33:04Now the other reason people use this

33:06particular extract feature is because they

33:08're only interested in a small part of the

33:09organization and other people use it to

33:13optimize the way they build their workbooks

33:14and want to make it go faster.

33:15They want to make it easier.

33:16So again that's a really great way of doing

33:18that.

33:19OK.

33:19So that is an extract in a nutshell.

33:22There's definitely more I could talk about

33:24with extracts.

33:25There's definitely more you can learn.

33:27Go ahead and Google the topic.

33:29Look on YouTube.

33:30I've done videos.

33:31Other people have done great videos on this

33:32.

33:33Go ahead and check them out in the recorded

33:35version I may be put up some links to some

33:37resources that I think are fantastic that I

33:38've not made other people have made.

33:40But I think you should check out.

33:41OK.

33:42Great stuff.

33:44So we've created an extract.

33:45We could do the same with this first one as

33:48well.

33:48And this is another way of creating an

33:50extract.

33:50You can just right click the file here.

33:52Select extract and you get the same window

33:55this time here rather than in the

33:57connection.

33:57You could add a filter in this case I went

34:00I'll go down to the bottom right select

34:02extract will ask me where I want to save it

34:04again.

34:05I hit my desktop save and you'll see that

34:08this extract has a different time zone to

34:10the underlying data.

34:12Since you table is looking at the metadata

34:14to understand hey what's going on here.

34:16I'm fine with this.

34:18I know this is fine.

34:19I'll select and show again.

34:20Click OK.

34:21And again we have an extract right there.

34:23So pretty easy pretty simple.

34:26Now at this point this is when I start to

34:28save the work.

34:29OK.

34:29I don't want to go through all this effort

34:31making connections and forget to save it.

34:33So if I click on Tablo and I go to file at

34:36the very top I can go ahead and save as now

34:39if I click save.

34:41It'll just go ahead and open this window

34:43and the one to save it in my Tablo

34:45repository.

34:46I don't want in my Tablo repository.

34:48I want it on my desktop.

34:49So let's go ahead and select my desktop and

34:51you'll see it gives it a name called Book 1

34:53.

34:53TWB I'll call it live stream.

34:56OK.

34:58Given it a file live stream now this file

35:00extension is super important.

35:02So many people make this mistake.

35:04So pay attention.

35:06OK.

35:06A TWB file is just known as a Tablo work

35:10book.

35:11At the bottom there's another type called a

35:14packaged workbook.

35:16OK.

35:17And the package workbook not only contains

35:19your data visualization but it also

35:22contains your extracts and any data sources

35:25as well as assets like icons and images.

35:27Those will get packaged into a TWB X file.

35:30And so most people I think generally want

35:33to save everything in one file.

35:36They don't want to save in lots of

35:37different places then have to go back and

35:39find it and relink it especially if you're

35:41about to delete the extract from our

35:42desktop.

35:43We just want to go back here and save

35:45everything in one file.

35:47So if I go ahead and select that we're

35:48going to save it as a TWB X file.

35:50It's safe.

35:51It will go ahead save it.

35:54I know that's safe because here at the top

35:56just the annotators kind of gone in the way

35:58here at the top you can see it's called

36:00live stream.

36:01And now that's saved.

36:03OK.

36:03And so what that allows me to do if I find

36:07my little icon is I can go ahead and I can

36:10now safely delete these two.

36:14Because they are actually now in my work

36:16book they've been saved as a package work

36:18book.

36:19If they weren't saved it would just save a

36:21TWB and the next time I try and go and do

36:24something it would still be connecting to

36:27the local files and it would basically have

36:29a panic attack because it's saying hey I

36:31had extract.

36:32They were on the desktop.

36:34Where have they gone.

36:35And so to fix that the only way you can

36:37kind of get those back is to go back

36:39connect to the original data source and

36:41regenerate the extract essentially this

36:43option here is a couple of options

36:44sometimes here you can use to regenerate

36:46the extract refresh re-update.

36:48Tableau will kind of give you a hint as to

36:50what it needs to do.

36:51But that's where you find these sections

36:53you can mess around with these options you

36:55know figure out what they do Google them

36:57whatever but we won't cover that in much

36:59detail.

36:59Go check out the videos on extracts that

37:01have been done by the community.

37:03OK.

37:04And now even though I've got my extract in

37:06my workbook I can go ahead right click on

37:08to use extract and it brings everything

37:10back in is what it now does as it goes to

37:12find the extract realizes it's not there

37:14then goes to the main file and brings all

37:16the data in.

37:17If I go ahead and extract the data this

37:20time removing my filter for subcategory and

37:23bringing everything in select extract we

37:26take the extract.

37:28You probably wondering why am I repeating

37:30extract because people don't get it.

37:33So just want to make absolutely crystal

37:35clear that you understand what extracts are

37:36.

37:36OK.

37:37The very final thing you can see a tick

37:39mark on the data source that I'm currently

37:42using.

37:42OK.

37:43I go to another data source you'll see that

37:45I don't get that tick mark because I haven

37:47't brought anything in and everything goes

37:49orange.

37:50If you're into that zone you're skipping

37:52way ahead.

37:52Just go back to the data source you were

37:55using.

37:55Select that.

37:56Clear the sheet.

37:58Just removing everything.

38:00Or I go back one step there's a back button

38:02right here.

38:03If I go back one step you can just go ahead

38:06here and select clear the sheets that will

38:08clear everything and now when I switch to

38:11another data source that blue stick isn't

38:13there and nothing's changing orange.

38:15OK.

38:15You just want to make sure you don't get

38:16confused.

38:17This is exactly what happens to people and

38:19they get so confused and they realize that

38:21creating blends and joins off gets crazy.

38:24So make sure you're working on the data

38:25source you're supposed to be working on by

38:27just making sure you clear the sheet or you

38:29create a new sheet if you need to.

38:31OK.

38:32So that is an extract.

38:35I've covered extracts in pretty pretty good

38:38depth.

38:38OK.

38:39Thank you.

38:44[ Silence ]