Tableau 2020.2: ESRI web data connector in Tableau

0:00In Tableau 2020.2 Tableau have added the

0:03ability to connect to your location data in

0:06Esri by simply

0:07entering an ArcGIS server URL or a Geo

0:09Service API URL. This makes it much much

0:12easier to connect to

0:14these data sources. So let's take a look

0:16and see how that works. The first thing you

0:17need to do is

0:18head to the server connection space here.

0:20You will notice that if you're on a Windows

0:23machine you'll

0:24have more of these connections than I do

0:25because I'm on a Mac, so don't worry about

0:27that if these

0:28are slightly different. You'll also notice

0:30that the web data connection option over

0:33here is

0:33available to you. This is ideally targeted

0:35at developers who are building their own

0:37connectors

0:38for use in Tableau but the one we actually

0:40want is called the Esri web data connector

0:43and you can see

0:43here that it's called Esri ArcGIS server

0:46and when you click on that you get this

0:48interface. Now in

0:49this interface you need to enter the server

0:52URL and if we do that what Tableau will

0:54then do once

0:55you've pressed connect is it will go off to

0:57the API and start querying data from that

1:00data source

1:01and you'll see that happen here in front of

1:03the screen depending on the quality of your

1:05connection

1:06it might take different lengths of time but

1:08just bear with the service and eventually

1:09you'll get

1:10a view like this one where you can actually

1:12see the individual fields that are

1:14available from the

1:15API and not only that if I hit update now

1:18you can actually see the information it's

1:21pulling through

1:21the metadata as it perceives it so the

1:24address information the city it's

1:25recognizing as geographic

1:27fields here if I just zoom in here and also

1:29you can see that the shape files are coming

1:31through

1:32as both polygons and multi-polygons so you

1:34've got different spatial types in a single

1:36column there

1:37so that's quite handy to be aware of okay

1:39and now once you've done this you can just

1:41go ahead and

1:42start visualizing with it as you would and

1:45if I hit on sheet one here that takes me to

1:48the next

1:48data source bear in mind it has to create

1:50an extract it can't connect to these APIs

1:53live

1:53because of the way it needs to be able to

1:55work with the data so it always creates

1:57this extract

1:58we know that because of this icon you can

2:00see here these two cylinders with an arrow

2:02between them

2:03that is the sign that lets you know that

2:05you're looking at a tableau server extract

2:08now that we've

2:08got our data source here we can start

2:10visualizing with it so I'll do a very very

2:12simple chart

2:13I'll double click on shape which we know is

2:15our spatial field that generates this map

2:17then I'll

2:18color these by the name of these boundaries

2:20and I'll add all the members tableau just

2:22letting me

2:23know there's a lot of data there and there

2:24we have it we have all the names of the

2:26different areas

2:27very easy to see in tableau we might make

2:29the map a little bit more appealing by

2:32changing the style

2:33to maybe an outdoor style and then turning

2:36the sort of washout down so we can still

2:38see the main

2:39bit of the information but of course as you

2:41zoom in you'll see that this map gets more

2:43and more

2:43detailed and you can actually start to sort

2:46of gain some context from the map so there

2:48you have

2:48it in literally no time we've built a very

2:51simple chart using spatial data and that

2:54connector now

2:55there is one more thing I'd love to show

2:57you and that is the fact that when you

2:58connect to any data

2:59source in tableau especially new ones you

3:02always get the capabilities of tableau up

3:05until that

3:06current version so with esri what we've got

3:08here is not just the ability to connect to

3:11esri data

3:11sources but actually we can add new

3:13connections so this time around I'm going

3:15to do something

3:16slightly different I want to do something

3:19called across database joins using this

3:22kind of data so

3:23the first thing we'll do is I've got a link

3:25I'm actually getting these links from the

3:28tableau

3:28beta for 2020.2 I'll post them in the

3:31description below so you can have a go

3:33along with this

3:34so look out for the links in scenario

3:36number two but essentially I want to create

3:38a new connection

3:39I'll go to esri arcgis again and this time

3:42I'm going to paste the first link here and

3:45we're just

3:45going to go ahead and connect to that and

3:47this is creating a new connection so you

3:49can see it's

3:50almost the same data as we had before and

3:52now what I want to do is add a second

3:55connection here so

3:56I'm just going to copy this link I've got

3:58it off screen here so you just have to

4:00trust what I'm

4:01doing I'm going to add a second connection

4:03here I'm just going to paste that here so

4:06you can see

4:06I'm pasting it to the server url space I'm

4:09hitting connect and now as it loads that

4:11information you'll

4:12notice two important things the view that I

4:15have here has this blue bar going across

4:18the top this

4:19signals that it's actually coming from the

4:21first connection and I know that because

4:23the colors

4:23match if I zoom in here on the top you'll

4:26see I have this blue bar on the left of

4:28this connection

4:29and that's actually the same as what I've

4:31got here and you can see the name of the

4:34table that

4:34it's coming from and that matches this

4:37table okay so what I'd like to do is I'd

4:40like to see which

4:41government buildings reside within a

4:44certain water service agency okay and now

4:46in the old way of

4:47Tableau what I do next is I just drag out

4:49government buildings and I'd leave it there

4:52but

4:52you see I get a warning and that's because

4:54Tableau thinks I'm trying to make a

4:56relationship you need

4:57to check out my video on relationships

4:59which have been released also in 2020.2 but

5:02essentially this

5:03is the wrong type of thing we're trying to

5:05do it would actually be easier if we could

5:07do a spatial

5:08intersect think of that as like a spatial

5:10join with the two data sets so let's remove

5:12this and

5:14head into this particular connection and

5:17now you see I get the traditional view for

5:19joining tables

5:20and now if I drag the government buildings

5:23in you'll actually get the interface to

5:25join those

5:25two data sources now the next thing we need

5:28to check is to make sure it's joining on

5:30the correct

5:30thing so you see here we've got city on the

5:33left and city on the right this doesn't

5:35make as much

5:36sense because what I actually want to do is

5:38use the spatial objects to make sure that

5:41the buildings

5:42reside in the city I don't want to just

5:44trust the city columns as being correct

5:46sometimes there's

5:47data errors out there so I'll use the shape

5:50object which I know is a spatial field and

5:53I'll use the

5:53shape object in my second data source so

5:56now I'm matching these two but you'll

5:58notice that I've got

5:59the equal sign here and this isn't actually

6:01the correct thing to do I shouldn't be

6:03checking if the

6:04two spatial objects match I should check

6:07whether they intersect and now that I've

6:10done that you can

6:11see here that I get the traditional options

6:14for joins inner and left I actually only

6:17want data

6:18that maps onto these service agencies okay

6:21so I'm just going to hit update now to see

6:23what the

6:23resulting view looks like and if you recall

6:26earlier on I mentioned that the colors

6:28matter

6:28you'll see orange is now showing here

6:30because it's brought in data from my

6:33secondary connection so

6:34here I am with a new connector in tableau

6:36to the Esri data source and I've done a

6:38cross database

6:39join using the new capabilities with data

6:42models and relationships as well baked into

6:45that so it's

6:46quite a mind-melter I appreciate there's a

6:48lot to learn here but check out my

6:49individual videos on

6:51those topics and we'll be able to get you

6:53up to speed now you can see here that all

6:55these areas

6:56are in Seattle so let's over head on over

6:58to the new sheet and let's start visual

7:01izing that

7:02remember it creates an extract every time

7:04you do this because it has to bring the

7:06data locally

7:07and query that from the API but now that we

7:10've done this we can just simply bring out

7:13something to

7:15name the area so if I bring the name of the

7:17area you'll see here's the Seattle public

7:20utilities

7:20and I can grab all my buildings and just

7:22put that next to it and now I can see all

7:24the buildings

7:25that sit within this water service district

7:28and so if I'm a planner I actually

7:30understand you know

7:32if I need to go get a problem solved I need

7:34to go to Seattle public utilities to solve

7:37any utilities

7:38problem that's going to be generated by

7:40these guys now you can obviously build way

7:43more interesting

7:44charts you've got the full capability of

7:45all your data sources here as well so you

7:48can obviously

7:49play around with both data sources I can go

7:51ahead to another sheet and actually just

7:53bring back the

7:54map that I had before and we can even look

7:56and see well where are the different

7:59buildings if I just

8:00uh if I just grab the shape file for the

8:03buildings which I believe is actually here

8:07and I just visualize that as well let's get

8:09rid of the first one you can see these are

8:12the actual

8:12buildings and if I just zoom out and I

8:14bring some fidelity back onto that map and

8:17then we can

8:18actually start to see some sort of richness

8:20in the context that's sort of being played

8:22out here okay

8:24thanks very much for watching I appreciate

8:25this was a long one if you found this video

8:28useful

8:28hit like and subscribe and if not drop a

8:30comment below let me know what you'd like

8:33to see in the

8:33near future thank you very much