New Features coming to Tableau 2022.1

0:00Hey, it's Tim here. I haven't made a video

0:01in a long, long while.

0:03Essentially, I've just been under the

0:04weather for the last couple of months.

0:06I've had colds, I've had flus, I've been

0:08busy. Life's taken over, but

0:0922.1 is about to come out, so I'm back with

0:13another video going through all the

0:15features

0:15that are coming soon in 21.3. In the coming

0:18weeks, we'll obviously start covering all

0:20the videos to

0:21cover all the new features in this release,

0:23but for now, let's have a look at what's

0:24available

0:25on the website. As I go throughout this

0:27video, there's going to be timestamps,

0:28sort of covering all the sections. After

0:30the overview, if you just want to jump to a

0:32specific

0:32section and find out more about that, go to

0:34the chapters on your phone or on desktop

0:36and skip

0:37ahead to that section. Or just go to the

0:38description, open that up, and there'll be

0:40timestamps there to jump ahead and skip to

0:42the various bits you're interested in.

0:45Okay, so here we are on the coming soon

0:46page. Tableau normally releases this with

0:49about

0:50a month to maybe a month and a half to go.

0:52This came out last week, so I think 22.1 is

0:58probably

0:58going to land towards the end of March or

1:00maybe early April. It can always shift a

1:02week or two

1:03here while marketing get their stuff in

1:05order. But also, if you head to Tableau

1:07Online, some

1:08instances of Tableau Online have already

1:10upgraded to the latest version. So actually

1:12, if you're

1:12using Tableau Online, some of these

1:14features on this page might already be

1:16available to you. But

1:17let's just take a quick high-level overview

1:20of what's coming in this particular release

1:23,

1:23and then we'll probably just go into each

1:25of them in a little bit more detail in the

1:27future,

1:27of course, but also in this video, just

1:29digging into what they are. So we have the

1:32workbook

1:32optimizer, a new search experience, ask

1:35data phrase builder. Ask Data has been

1:37getting a lot

1:38of love every single release, so it's no

1:39surprise that that's still here. Enterprise

1:42deployment

1:42guidelines, background resource limits,

1:45customizing view data, data freshness,

1:48data freshness parquet file support. So it

1:51's always interesting. Tableau always put

1:53these

1:54capabilities in sort of an order and in

1:57grouping. So data freshness is the category

1:59,

2:00and that is essentially relating to a new

2:02connector that you can connect to. So that

2:05's what

2:05that is. We'll go into this in a little bit

2:07more detail. New accelerators on the Table

2:09au Exchange.

2:09The Tableau Exchange has been around with

2:11us now for a release. I was pretty critical

2:14of this in

2:14my video. I kind of watched that back and I

2:16thought, man, it really went in on that

2:18thing.

2:19But nonetheless, hopefully there's an

2:21opportunity for Tableau to improve over

2:23time.

2:24New connectors on the Tableau Exchange. We

2:26'll get into that. Tableau server

2:27independent gateways,

2:29hyperspooling monitoring, slack integration

2:32for monitoring, swap with root table for

2:34the data

2:35model, tabpy model evaluation function, so

2:38a new function for tabpy or specifically

2:41Python

2:41functions, filter performance optimization,

2:44improvements in Esri web data connector,

2:46so essentially the way you connect to Esri,

2:49which is a mapping capability that's

2:51independent of

2:52Tableau. But Tableau works pretty nicely

2:54with it and Tableau is improving that

2:55relationship.

2:56And a new toolbar for Visas, a web altering

2:58improvement. So web altering improvements

3:01is like a catchall for web edit getting

3:04better. So in essence, web edit is behind

3:08desktop editing

3:09experience. And so when they say web or

3:11train improvements, what they really mean

3:12is we're

3:13playing catch up with the desktop

3:15experience. And the idea is that eventually

3:17it will reach parity

3:18or it will be better than desktop

3:20experience, in which case at that point the

3:22web and desktop

3:24are pretty much on par and therefore it's

3:26easier for Tableau to just deploy Tableau

3:28through the

3:28browser, not having to wait for software

3:30installs and so on and so forth. So again,

3:32we'll get into

3:32this in more detail. View acceleration,

3:35automated identity migration, usage based

3:38publishing,

3:39frequency for admin insights, linked tasks

3:41with prep conductor. I believe this is an

3:43enhancement

3:44to the previous release of this. Vis load

3:47times, data source for admin insights. So

3:50essentially

3:50a new data source that talks about Vis load

3:52times. Parameter enhancements in Tableau

3:55prep,

3:55customer data platform accelerator in Table

3:58au online, prep flow notifications in Slack,

4:00virtual connection support for Salesforce

4:02data. So essentially the ability to create

4:04virtual connections for Salesforce data

4:06sources, which is super interesting. I'll

4:08come back to

4:08that in a second. Improved OAuth login

4:10processes, virtual connections integrated

4:13with the Tableau

4:13catalog. And lastly, the embedded API

4:17version 3.1. So the .1 release is normally,

4:22I call it

4:22sort of the spring release. I sort of think

4:25of this as the release where Tableau put in

4:27all the

4:27groundwork for the awesome stuff they're

4:30about to do throughout the year. So you

4:32might go through

4:33this list and it might just mean absolutely

4:34nothing to you. It might mean like, you

4:36know,

4:37all this stuff is just little tidbits. You

4:39probably will only have an affinity with

4:41two or three of

4:42these features if you use Tableau every day

4:43. If you're an admin, you'll love some of

4:45these. If

4:46you're an author, you'll love some of these

4:47. If you're just a plain old user, you'll

4:49love some

4:49of these. But none of these are sort of big

4:51hitter sort of wow features. There is a

4:53couple, but none

4:54of them really sort of stand out on the

4:56same part of something like the data model.

4:59And the reason

4:59that is, is because again, it's

5:01foundational work. These are things they

5:03have to sort of get right

5:04before they can really bring out the

5:05features. It's normally the .2 or .3

5:07release that really come

5:09out with the headline features, the

5:10features that sort of carry the whole

5:12release, if that makes

5:13sense. Now let's get started with this page

5:15. The first feature is the Workbook Optim

5:17izer. And what

5:17I love about this page is it has GIFs of

5:19how things work. Now the Workbook Optimizer

5:22is not

5:23sort of as it sounds. It's not going to go

5:25in there and optimize your workbooks, but

5:26what it will do

5:27is it will show you things you're doing

5:29that are not necessarily based on the

5:31optimum experience.

5:33So you can see here a little GIF on the

5:34right hand side. You've got a little pop-up

5:36, a modal that

5:37turns up, and it basically gives you a

5:39bunch of things to check. It does 12 checks

5:42initially at

5:43launch, I believe, but it will do more in

5:44the future. That's something that Tableau

5:46have said it

5:47will improve on. And essentially the checks

5:51are basically checking for best practice

5:54guidelines

5:55around Tableau dashboarding. The whole

5:57point of this is to try and surface

5:59knowledge about

6:00building dashboards that's not necessarily

6:02apparent. If you've learned Tableau, maybe

6:05self-taught or if you've just learned on

6:07the job or if you've maybe just picked up

6:09what you've seen

6:09elsewhere, or maybe you're using things

6:11that you think are awesome, you've seen

6:12them on Tableau

6:13Public, but you're maybe not using them in

6:15the right place. The Workbook Optimizer is

6:17going to

6:18try and give you some insights as to how

6:19you can improve those sort of sticking

6:21points and make

6:22the experience for the end user better. And

6:24this starts off with things like

6:26performance, but it

6:27also will eventually in the future move to

6:29things like accessibility guidelines. So

6:31that's something

6:32that's really sort of powerful. And what I

6:34like about this is it actually gives you

6:36feedback. So

6:37it checks your workbook, it looks at very

6:39basic things, you know, the number of rows

6:43you may be

6:43displaying all at once. If you've got big

6:45tables that will obviously come up here as

6:47a sort of a

6:48warning, but it's also sort of got a red,

6:50amber, green indicator. I mean, Tableau

6:52doesn't typically

6:53stick to those colors, but here we go. We

6:56've got red, amber and green indicators

6:59telling you sort

7:00of what's going on. And each of those gives

7:02you an action. It says, hey, look, the work

7:05book contains

7:06this number of sheets. The number of sheets

7:08will do this to your workbook. And then you

7:10can say,

7:10well, maybe I can build this in one sheet.

7:12Maybe I can do this in two sheets. Maybe I

7:13can use

7:14less sheets because the user is not

7:16actually using half of these sheets. And it

7:18also tells you what

7:19you have done well. So it doesn't just tell

7:21you what you've done badly. It tells you

7:22what you've

7:22done well. So I think this is a really nice

7:26feature. What I hope this does is not only

7:28start to surface this for authors, but I'd

7:31love this to become an API so that admins

7:34can go in,

7:34check the whole server, pull out the work

7:36books that score really poorly, maybe set a

7:38threshold

7:39and start to use this as a way of helping

7:42people progress through governance in an

7:45organization in

7:46a sort of much more structured way. But I

7:48think it needs to have all the tests and

7:50capability built

7:51in. And at the moment, it's still at launch

7:5312 checks. I'm hoping this sort of goes out

7:55to more.

7:56And I'm also hoping you get control about

7:58which checks it does and how they all work.

8:00So I'm

8:01really looking forward to see how this

8:03lands, especially when people start getting

8:05to this.

8:05I'm pretty sure this is already available

8:07on Tableau online, so go check it out there

8:09.

8:09But otherwise, check back in a week or two

8:11when I make a video on this feature.

8:13Now, the new search experience now search

8:15is one of those things where you don't

8:17really know it's

8:17better until you've been using Tableau for

8:19a while and then they change the experience

8:21for

8:22everyone who's starting Tableau today, they

8:24'll never know the search wasn't great. This

8:26new

8:26search feature, I think is going to work a

8:28lot better. I've seen this in demos, I've

8:30seen a

8:30couple of demo this and I genuinely think

8:32that it is going to be better. It's one of

8:34those things

8:35where you can't really know for sure how

8:37well it works until you actually need to

8:39find something.

8:40And you go in you search and the first few

8:42things that it shows you are actually what

8:44you're looking

8:45for. And what they've done is essentially

8:47changed the algorithm behind the way it

8:49searches around

8:50specific criteria. So it's it's quite

8:52intelligent. I'm not going to bore you with

8:54all the details.

8:55I'll do this in my full explainer video,

8:57you know how I understand this. But

8:59nonetheless,

9:00and the search experience when people go to

9:01the top right and search something,

9:03it's going to be much improved. It's going

9:05to have a much better hierarchy. And you

9:06can see here is

9:07even got the ability to filter things in, I

9:09think, a clearer way, something that you're

9:11more familiar

9:12with saying on the web today than maybe

9:13what we had in the past, which was a little

9:15bit clunky in

9:16places. So this is going to be a nice touch

9:18. The R-Stata Phrase Builder. Now, this one

9:21's interesting

9:22because R-Stata has sort of been improving.

9:23It's one of those things. It's one of those

9:25features,

9:26actually, that when you come out with

9:28version one, version one is not what you

9:30intend to actually

9:31ship. Version one is very much a journey

9:33that you sort of embark on with your

9:34customers.

9:35Now, more recently, I'd sort of say we got

9:37version two of R-Stata. It started to have

9:39a much,

9:40much more well thought through interface,

9:42an interface that sort of helps you build

9:45questions. But more importantly, they added

9:47the ability to only let you use R-Stata

9:49where a lens

9:50has been created. And the lens forces a

9:53sort of specificity around the question

9:55that you're trying

9:56to answer. And because of that, the

9:57questions are going to be better. So if I

9:59create a lens

10:00about shipping data, then I'm only going to

10:02be looking at shipping data in my sort of

10:04order

10:05data set. What they're doing here is

10:07actually adding to that capability of

10:09building phrases.

10:11So what you can essentially do is use this

10:13new feature to build different types of

10:15questions.

10:15They're essentially sort of guiding you,

10:17the user, through how to use it. And what I

10:20think

10:20will happen over time is as you start to

10:22understand the way it works, you will start

10:24to understand

10:25how it expects these questions to be asked,

10:27but also how to get stuff out of it.

10:29In every single application, in every

10:31single tool, there's always like a hack. I

10:33'm going to call it a

10:33hack. But there's always sort of a way of

10:35doing things that works really well with

10:37the application.

10:38And once you know it, it's amazing. It's a

10:40bit like Gmail and search. You can just go

10:43ahead and

10:43type what you're searching for and it will

10:44find it. But you can actually add search

10:46parameters

10:46that can filter things down. The same in

10:48Outlook, same in other tools. This is the

10:50kind of capability

10:51that's sort of building into our state here

10:53. So I think this is going to be fantastic

10:55for those

10:55people who are using this. But again, if

10:56you're an author, you probably don't use

10:58this. You use the

10:59power tools, so it might not be a big thing

11:01for you. Okay. Next one is enterprise

11:03deployment

11:04guidelines. So this is essentially a change

11:07in the guidelines around how to deploy

11:09Tableau.

11:10And it's been sourced by basically looking

11:13at infrastructure all the way down to the,

11:16you know, they call it bare metal. I'm not

11:18a developer, but they call it bare metal

11:20because

11:20they're really literally looking at how the

11:22software performs on the infrastructure

11:24that

11:25most people run, the actual silicon itself.

11:27And so what they've come up with is a

11:29improved set of

11:30enterprise deployment guidelines that give

11:32a standardized spec and they give a

11:34standardized

11:35route for deployment. And this is important

11:37because companies spend a lot of time and

11:39effort

11:39sort of tweaking their servers, trying to

11:41get the perfect sort of alignment of

11:43features. And so what

11:44this should do is just highlight, you know,

11:47look, this is the setup that we've tested

11:49and we know

11:50works. So just go get this, install it. And

11:52also when things start to go wrong, if you

11:54follow these

11:54guidelines, Tableau are more likely to

11:56understand what's going on. Whereas I think

11:58a lot of support

11:59requests today, Tableau will spend a lot of

12:01time and energy trying to understand your

12:03setup first,

12:04and then they can start to help you. So I

12:06think this will be really nice. I've

12:08probably maligned

12:08this a little bit. So be sure to sort of

12:10check my video out when I go into this in

12:12more depth.

12:13Okay, if we go to the next one around

12:14background resource limits, essentially,

12:16this is just the

12:16ability to control the background. The

12:19background is a service that does jobs on

12:21Tableau server.

12:22And with this, you can essentially control

12:25how many jobs and sort of how much of the

12:27background

12:28is a particular site is taking up. So here

12:30on the screen, you can see, let me just get

12:32my,

12:32I haven't even got my annotated, man, you

12:34can tell I haven't done the video in a

12:36while because

12:36I don't even have my annotator running. I

12:38normally always have that. So if we get my

12:40annotator out,

12:41you can see here we have a list of sites.

12:44And these sites in most enterprise

12:46organizations,

12:47you have more than one site, a site is

12:48essentially like a part of the organization

12:50in Tableau,

12:51so to speak. And but here on the right hand

12:54side, you have a bunch of resource limits.

12:56So you have

12:57the subscriptions, extracts, flows, and so

13:00on and so forth. And what you can do is you

13:02can

13:02essentially set a limit. So you can

13:05actually say, look, this is the current set

13:07of defaults that are

13:08going to be applied across the board. And

13:10you can set these and you can sort of apply

13:12them. But more

13:13importantly, you get like a nice visual

13:15table display of you know how this is

13:18working. The

13:19reason this is important is because in some

13:21businesses and some organizations, they

13:23split

13:24up the server into different organizational

13:26units and sites. The reason that is, is,

13:29you know,

13:29for legal reasons, you might not be able to

13:31, for example, say have HR data available to

13:33everyone in

13:34the organization. So the way you solve that

13:36is you put it on a separate site, and you

13:37give only

13:38the people who have access to that data

13:40access to that site. And what you want to

13:42do is you make

13:43sure that you can, at an overall level,

13:45guarantee a good level of service to

13:47everyone, not just one

13:48particular team. So this allows you to sort

13:50of standardize that. So if one more part of

13:52the

13:52organization wants to run away with

13:54background resources, you can't really stop

13:56them at the

13:57moment other than sort of phoning them and

13:59telling them, hey, you need to sort of pare

14:01down the number

14:01of background as you guys are running. And

14:03at the moment, that's not sort of possible.

14:06But now this

14:06will be once this sort of rolls out. So

14:08that would be a really nice feature for

14:10admins and people who

14:11use server on premise. For Tableau Online,

14:13this isn't a problem because you don't need

14:15to manage

14:15this. So if you're a Tableau Online user,

14:17this is sort of non-consequential. But if

14:20you're a Tableau

14:20Server user, this is going to be great. If

14:22we move over to Customize View Data now, oh

14:24my God, this

14:25is such a good feature. This is probably

14:27one of those features where you don't

14:29understand how good

14:30it is until you use it. Or if you start

14:32learning Tableau today and you see this,

14:33you think this has

14:34always been part of the software and it's

14:36not. This is a feature that I'd consider

14:39low-hanging

14:39fruit. It's been an idea in the Ideas Forum

14:43for absolutely ages. But finally, finally,

14:46Tableau are

14:47going to be improving the way that you can

14:49view data in Tableau. I know that sounds

14:51crazy, but

14:52if you ever go to a data point, right-click

14:55it, open it up and view the data, it shows

14:57you a table.

14:58And that table is pretty much standard. It

15:00has one format and one design. What they're

15:02doing is

15:03enhancing that. So not only can you choose

15:05which columns show, you can change their

15:07order, you can

15:08even export different tables in a specific

15:10way. And the great thing about this is it

15:13gives a lot

15:13more customization and they've also gone a

15:15little bit further and they've made sure

15:17that it's

15:18compatible with accessibility guidelines

15:19that Tableau already supports. So it's a

15:22really nice

15:22feature. So I'm not going to sort of murder

15:24this too much now because it's the kind of

15:26thing I just

15:27have to show you, but this is going to be

15:29really, really nice. If we go to the next

15:31one, this is

15:32essentially a new connector labeled Data

15:34Freshness Parquet. This is the ability to

15:37essentially

15:38connect to, I believe, let me just double

15:40read this. So build solutions that connect

15:43Tableau much

15:43more easily and efficiently with a variety

15:45of data sources and freshness technologies

15:47that use Tableau.

15:48This is a lot of jargon. You can now use

15:50the REST API to ingest data from Parquet

15:53into Tableau

15:53Online and Tableau Server. So this is where

15:56I'm like, okay, what is Parquet? I think I

15:58know what

15:58it is. Let's say file format. Let's just do

16:03that and see what comes up. Data bricks. I

16:12wanted to

16:12say that it had something to do with AWS,

16:15but now why is AWS coming into my mind? Let

16:18me do this.

16:20Let's say Parquet and let's just do AWS.

16:22Let's see where am I getting this

16:23association. Okay, so

16:27I'm sorry, Parquet writes okay. So what I'm

16:30doing here, I'm just trying to understand

16:33where did I

16:33get, let's just search Parquet and see what

16:36comes up. There's also a place apparently.

16:40Apache Parquet

16:41documentation. Okay. Let's go to this. This

16:45seems to be the thing that keeps coming up.

16:48You know what? I'm going to take a humble

16:51note here and say I'm not entirely sure

16:54because

16:54I wanted to say it had something to do with

16:58files in S3, but that doesn't seem to be

17:01what it is

17:03unless I may be confusing services in AWS

17:04with this particular feature. But

17:06nonetheless,

17:07come back to a future video to find out

17:10more. Sometimes you don't know. You just

17:13don't know.

17:14Right, new accelerators on Tableau Exchange

17:16. Okay, so this is essentially what I think

17:19is going to be

17:20now the march of new capabilities on the

17:23Tableau Exchange. The accelerators are

17:25essentially

17:26dashboards. They're templates that you can

17:27download and get started without having to

17:29do stuff.

17:30Now, I was really critical of this part of

17:33the extension gallery. Now, if I right

17:36click on some

17:36of these links, I think some of these have

17:38actually been already done and released. So

17:41we can go check

17:42them out. I'm going to open all of them. I

17:44probably could have done this much, much

17:46faster, but I'll

17:47just do it this way. Okay, so we've opened

17:49all of them. They're all in tabs on nature.

17:51They're all

17:51loaded. Now let's go check them out. So the

17:54first one, email marketing campaigns. Okay,

17:56we've got a

17:57little description and we've got a

18:00dashboard. Cool. Here we go. Banking loans.

18:04So this is like a loan

18:05dashboard about loans, essentially. Okay.

18:08Fraudulent claims. So fraud detection.

18:13And you remember, these are ticking

18:16specific boxes. Okay, one of the links didn

18:18't work. It's not a good

18:18start. If you're going to sort of promote

18:21new ones, then I guess make sure the links

18:24work.

18:24But nonetheless, hopefully that will...

18:27there might just be that because 21... well

18:29, 21.2 is definitely

18:31not out. Oh wait, 21.2 is out, so it should

18:35be fine. But let's see if we remove the

18:37version. Does

18:38this do anything different? No, it doesn't.

18:41Okay, no worries. I don't know what

18:44actually triggered

18:44that. So 1234, the fourth link, 1234 sales,

18:48advance plus target is what triggered that.

18:50So that's a

18:51broken link. We'll come back to that later.

18:54Insurance claims. Here we go. Insurance

18:57dashboards.

18:57We've got a bunch of charts there. Pretty

18:59cool. I'm not going through each of the

19:02tabs that I would

19:02take forever. I'm just looking at the

19:05landing page and seeing what we see. Okay,

19:07accounts receivable.

19:09Okay, yeah, lots of tables. Call center

19:13expert. So the call center monitoring. Okay

19:17, so, you know,

19:19I still stand by what I said. Look, these

19:21are... these are average dashboards. They

19:24don't have any

19:24styling. They don't have any formatting.

19:26They're essentially supposed to be vanilla,

19:28I think,

19:28to meet the needs of many. So you can go in

19:30there, get ideas, steal them, get started,

19:33immediately

19:34create some value as soon as you get into

19:36Tableau. Is it what people are building

19:38best practice in

19:39Tableau? No. Is it what the community is

19:41sort of knocking out on Tableau Public? No,

19:42it's not.

19:43Last time I did a showcase where I went to

19:45Tableau Public and I searched the same

19:48topics

19:48and found better examples of dashboards. So

19:51I think the jury's still out on that. But

19:53these

19:53things take time. You know, when you look

19:55at Apple and the App Store, that the first

19:57version of the

19:57App Store was diabolical and it had a lot

19:59of teething problems. And essentially, you

20:02have to

20:02figure these things out. You have to figure

20:03out what works by just putting something

20:05out there

20:06and seeing what people pick up. So I'm not

20:08going to critique that aspect of it. At

20:10least we're

20:10getting visibility. At least we're seeing

20:12updates. And that's a good thing. Right.

20:15New connectors on

20:16Tableau Exchange. Now, these are going to

20:19be interesting because I believe that these

20:22are

20:22probably going to be made by third parties

20:24because the Exchange is obviously not for

20:26Tableau to build

20:27extensions. Tableau could put stuff on

20:29there. But I think the main emphasis here

20:31is for third parties

20:32to build solutions. And you can see here

20:35they actually have and they're building

20:37things that

20:37are backwards compatible. So Tableau 2020.3

20:41and later will support this. And if I go

20:43back,

20:43Amazon DocumentDB is built by AWS, which is

20:46pretty interesting. So this might is this

20:50what I was

20:50thinking? Is this what I was confusing? I

20:52don't know. I'm not going to try and

20:53pretend I know the

20:54answer to this. But again, it goes back to

20:562020.3 and the developer, this is going to

20:59send you to

21:00AWS. So if you click on that, you'll go to

21:03this page. And this is the AWS page for

21:06this particular

21:08feature. And I think this this must have

21:11been written. Oh, yeah. So this is written

21:16for lots of

21:17different tools. So the JDBC drive for the

21:20Amazon DocumentDB managed a database

21:22provider on SQL

21:23racial for developers and BI tool users. So

21:27this is going to serve for multiple tools.

21:30It just so

21:31happens that I think Tableau have enabled

21:33the capability that allows AWS to turn up

21:35and just

21:35put their connector right there. So this is

21:37actually a great example of the exchange

21:39working.

21:39AWS, a pretty big player in the data space

21:43can just open up a couple of developers to

21:47change

21:47some code and some of the tools and boom,

21:50we have a connector for an AWS capability.

21:54And

21:54alternative connector for ClickHouse. I don

21:56't know what ClickHouse is, to be honest, so

21:58I'm not going

21:59to pretend to know. But hey, we've got a

22:01taco file. The taco file is the connect is

22:03the file

22:03format, you can see here that you've got a

22:06dot taco, that is the file format here. So

22:08that's why

22:08I called it a taco file. But nonetheless,

22:11the new connector, that's great. Let me

22:13close all of these

22:14tabs. Let's get these out of the way. And I

22:17go back to the main page, right. So we are

22:20kind of how far

22:21are we we're not even halfway. So we've got

22:24to pick up pace here. The Tableau server

22:26independent

22:27gateway. It's all the gateway process

22:29outside of Tableau server. So when you

22:31install the Tableau

22:32server, there are processes. In fact, if I

22:34get my annotation tools out, you can see

22:36that these are

22:36the processes. Now each of these processes

22:38is responsible for doing something. I know

22:40this

22:41because I'm about to start going through my

22:43Tableau server certification process again.

22:46So

22:46in essence, each of these processes does

22:48something different. One of the key parts

22:51of the processes

22:53used to be the gateway and the gateway

22:55could only live inside of the Tableau

22:56platform. What

22:57essentially they've done is they've allowed

22:59you to install an independent gateway

23:01essentially to

23:02allow you to do this elsewhere. So you can

23:04see here that the gateway which I highlight

23:06here going

23:06across is external. So gateway external

23:11means we also have a new process. Now this

23:14is interesting

23:15because I'm sure this is not the first,

23:18well this won't be the last sorry, process

23:21that Tableau will

23:22allow you to externalize. Things like

23:25taking backups would be really useful if

23:27for example

23:28you could put your repository in an

23:30external place. A repository is just where

23:32Tableau keeps

23:33its stuff. If you could put the repository

23:35elsewhere that would be really really cool.

23:38So again

23:38watch the space, let's see what happens

23:40here. But my hunch is that some of these

23:43will also

23:44end up being external. Hyperspooling

23:47monitoring. So I don't know what this means

23:50so let's have a read.

23:50Understand when hyper writes to disk with

23:52Tableau server management the resource

23:54monitoring tool

23:56will now also monitor hyperspooling. So I

23:58don't know what it is. Configure hyperspool

24:00ing incidents

24:01to be notified when there is not enough

24:03memory available. So I think hyperspooling

24:05is essentially

24:06just the process of hyper being written to

24:08a disk. I think that's what it means. Let's

24:12keep reading

24:13here though. Configure hyperspooling

24:14incidents to be notified when there is not

24:16enough available

24:17memory and the Tableau server data engine

24:19process shifts to spooling which impacts

24:21performance. So

24:26what I'm figuring here is hyperspooling is

24:28not good and they're giving you a way to

24:30monitor it.

24:31That's the most basic understanding I can

24:34figure out. So yeah hyperspooling comes up

24:38with the

24:38Tableau server data engine. So let's go in

24:41here and search for the term of spooling.

24:44This is by

24:45the way the way I do research. If I don't

24:46know something I go google, I put the

24:48phrase in. Google

24:49helps me out because it knows what I like.

24:51I go to the documentation and get the

24:53documentation has

24:54an answer. So let's read this. Memory usage

24:57. Okay in most cases systems have enough

24:59memory to do

25:00these types of processing but if there isn

25:02't enough available memory or more than 80%

25:04of RAM

25:04is utilized the data engine shifts to spool

25:07ing by temporarily writing to disk. Man that

25:10took me

25:11forever to get to. I know what that is. I

25:13know what spooling is. I just didn't know

25:15what it was

25:15called. So the best way to think of it is

25:18your computer has lots of different ways of

25:21storing

25:21information. I'll go back to this page. The

25:24most common one we know is a hard drive. So

25:27that's for

25:28storing information when your computer

25:30switches off. When you switch it back on

25:32the information

25:32is still on the hard drive. That's sort of

25:35long-term memory. The next type of memory

25:38is

25:38something called RAM, random access memory.

25:40That is essentially the memory that the

25:42computer, the CPU

25:44can use instantly. When you switch your

25:46computer off this does not stick around.

25:48What this feature

25:49is essentially saying is that it's going to

25:51allow you to monitor when Tableau has to

25:53start writing

25:53to disk because it's run out of that fast

25:56access memory. That fast access memory runs

25:59really quick.

26:00We're talking you know terabytes. Well on a

26:03server you'd hope terabytes per second. The

26:06bandwidth

26:06between the CPU and the RAM is at that

26:09speed but the actual information goes back

26:12and forth very

26:14quickly which is why the bandwidth has to

26:15be really really high. Tableau is going to

26:17allow you

26:17to monitor when this is happening so you

26:19can start to optimize your server in sort

26:21of much much better

26:22way. So this is good. It's essentially more

26:24visibility for admins and hopefully my

26:26explanation was okay

26:27there. Slack integration monitoring. So as

26:31part of Tableau server management post

26:34notifications

26:35about the health of Tableau server to Slack

26:37. This makes sense. Slack has integration

26:39with Tableau

26:40now. It's part of the Salesforce family. Of

26:42course you'd get notifications in Slack if

26:44something's

26:44going wrong. The catch with this all is

26:46Tableau server management add-on. You're

26:48going to have

26:49to have that in order for this to work. So

26:51yes the price of Tableau I think is going

26:52up a little bit.

26:53They keep adding add-ons and keep putting

26:56stuff away in them so the vanilla Tableau

26:58doesn't come

26:58with all of this stuff but again that's

27:00something maybe another video for. Swapping

27:03root tables. Now

27:04this is for authors. So this is essentially

27:06the ability to choose which table becomes

27:09your root

27:09table. Let's say you get stuck and you

27:11realize you've built the data model up in

27:13the wrong way.

27:14The only solution you had was to rebuild

27:16the whole entire model and what you can do

27:18with this is you

27:18can just switch what the root model is and

27:20you're going to be away to the races. So

27:22this is really

27:23nice. Nice to see and I'm going to love it.

27:26Tab pi model evaluation functions. So this

27:30is interesting.

27:31I don't understand enough about Python to

27:33really know what this is off the top of my

27:35head. Let's

27:35have a look. Call model extension x has

27:38been added to Tableau specifically to call

27:40deployed models

27:41from Tab pi or other analytics extension

27:43that gives administrators a new

27:45configuration option

27:46to control Tab pi developments by disabling

27:49the execution of arbitrary Python code.

27:51Administrators

27:53can now maintain a single source of truth

27:55in analytics extension server by leveraging

27:57a central

27:58version control. Interesting. I still don't

28:03know what this means. Improved experience

28:07for ACX is

28:08deployed function in the Tab pi server. I

28:09've seen the model requirements before

28:11adding it to Tableau

28:12desktop. I think this might be a way for

28:15people who are using Tab pi to essentially

28:18evaluate

28:20whether a model they're using for Tableau

28:23or in Python is actually going to work for

28:26the use case

28:26they have without first having to bring in

28:29the function, do it, and then realize the

28:32calculations

28:32are wrong if that makes sense. So I think

28:34this is good. Let's look at this GIF. So

28:37you have model

28:38and extension and they've got real. And

28:41this is someone essentially testing this I

28:44think.

28:44Profit sales boom boom and they can go and

28:47test it. Okay I think this is what this is

28:51doing.

28:51Call deployed models from Tab pi or other

28:55extensions. So this is the key thing. This

28:58is

28:58all that's happening here. I think what I

29:00don't understand is how is this different

29:02to what you

29:03could already do and maybe if you know let

29:05me know in the comments. I don't know so I

29:07'm not

29:08going to pretend to know. Find out in a

29:10future video. This is happening a lot this

29:12time around.

29:12Filter performance optimization. Okay so

29:15this essentially I think just the

29:18improvement of

29:19the filter pane. So Tableau 2021 introduces

29:21filter performance at scale. Whether your

29:24dashboard has

29:24one, eight, or more visualization controls

29:26you will have a highly responsive and

29:28consistent

29:29experience. Enhancement in filter

29:32performance will deliver a much faster

29:34experience when using

29:36show filters. Session cloning, filter item

29:38selection, mark selection, tab switching,

29:40show me screen. Okay so when you have lots

29:43of filters in a dashboard it does slow down

29:45.

29:45What they're saying is that shouldn't

29:47happen as much anymore. I still think there

29:49's a limit

29:49but eight is an unreal number of filters

29:53but I come across many dashboards with more

29:56than eight

29:56so they clearly have to do something to

29:58sort of bridge that gap and I think they've

30:01done the work

30:01to do that. So let's have a look at the

30:03next one. Improvements in Esri web data

30:05connector. So connect

30:06Tableau to authenticated ArcGIS servers to

30:09extract location data from Esri. So I think

30:12it's essentially

30:12just the improvement of the connector here.

30:14So it's another connector. I would put this

30:16under data and

30:17freshness but it's not new because it's an

30:19improvement to the existing connector so

30:21that's

30:21why it's a separate one. The new viz

30:23toolbar. Now this to me is a classic

30:26example of improving

30:28something because you're about to change

30:30something else. Okay so the new viz toolbar

30:33this is in

30:33change to the interface. It's going across

30:35the top and what they've done is

30:36essentially they've

30:37grouped certain activities. You can see the

30:40prominent aspects to the left hand side and

30:42then

30:42they've got this new action section called

30:44keep track which has subscription metrics

30:47and alerts.

30:47Then you've got exporting which has its own

30:49things, sharing which has its own things,

30:52and then also the ability to download data

30:55and a bunch of other things. Commenting

30:57which is

30:57probably going to ping stuff off elsewhere.

31:00So the top level groupings make a lot more

31:03sense and I

31:03think it's really really well done here. So

31:05yeah props to Tableau for this and what I'm

31:08interested

31:08in is what it will enable in the future and

31:10how that's going to work. So that's going

31:11to be really

31:12cool. Web altering improvements. So actions

31:15right where you want them. So actions and

31:18format analytic

31:18lines. Okay so formatting improvements and

31:20actions are basically getting improved. I'm

31:23not going to

31:23dwell on this because this is normally

31:25stuff we already have. Web edit is just

31:27getting them in

31:27this particular release. View acceleration.

31:30So this is interesting. Make workbooks up

31:33to

31:33100 times faster with the push of a button.

31:36Now this is bold. This is really bold. Work

31:39book owners

31:40can use the view acceleration button to

31:42accelerate their most important views that

31:44could use a boost

31:45in Tableau Online and Tableau Server. So

31:47there's a lot of questions about this but

31:50all you need to

31:51know as an end user is your admin can make

31:53things faster if they switch this on. What

31:56I'm interested

31:57now as an admin and as a sort of advanced

31:59user is what does that actually mean. There

32:02could be

32:02a workbook which doesn't have 100x

32:05performance to be gained and so is this

32:08just essentially taking

32:09a cache of a dashboard and instantiating it

32:11for longer periods of time or is it

32:13something more

32:14intelligent than that? Is it sort of

32:16dynamically looking at what people look at

32:17and then pre-loading

32:18that in the background so it kind of

32:20guesses your next step? And Google does

32:22this actually. When you

32:23go to Google and you hit a search page in

32:25the background depending on the settings

32:27your browser

32:27has it will pre-load and fetch the first

32:30three hits because the chance of you

32:32clicking on the

32:33first three hits is quite high. So by

32:36already going off and getting those pages

32:39when you then

32:40click on those things you get an

32:41instantaneous response and it's a really

32:43really fast experience.

32:44That's how you perceive the Google search

32:47engine to be fast. It's not because Google

32:49is fast it's

32:50because they're doing things in the

32:51background to give you the impression that

32:53it's fast. Same with

32:54this. So we'll hopefully see how that works

32:56when we get into it a little bit more.

32:58Automated identity

33:00migration. So this I think is about moving

33:03identity management from one system to

33:05another. We can see

33:06a screenshot here where we can see I think

33:10we're moving from one domain to another and

33:14yeah so

33:15Tableau has released a silent identity

33:18migration job which will move current

33:21system users to new

33:22identity service tables. This job will not

33:24interrupt or interface with any usage of

33:26Tableau and also

33:27net new installations to operate the new

33:29identity service by default seamlessly

33:31transition to a more

33:32modern identity data structure that

33:34enhances your data within the Tableau

33:37server repository

33:38with more information about the user's

33:40identity. Interesting. I don't know what's

33:43new here and it's

33:44probably because I'm a bit out of touch

33:45with the server features so I'll dive into

33:46that in the

33:47future. Usage-based publishing frequency

33:50for admin insights. So what's this? Usage-

33:53based publishing

33:54frequency and admin insights prioritizes

33:57updating the data in the content that's

33:59consumed,

34:00making data updates quicker in the sites

34:03that are used. For sites with low content

34:05usage,

34:06admins have the flexibility to choose if

34:08the admin insights content is published

34:10daily or weekly. So

34:11admin insights is a data source

34:15specifically for I think Tableau online

34:19although

34:21Tableau server users get it elsewhere in a

34:23different format. Essentially though,

34:26you can essentially change the publishing

34:29frequency for this data set. So make data

34:33up to quicker in the sites that are used

34:35for sites with low content usage having the

34:37flexibility to

34:37choose if the admin insights content is

34:39published daily or weekly. So a little bit

34:42more flexibility.

34:43I think you have to see this in practice

34:45because choice doesn't always mean you have

34:48the most

34:48flexibility available. You could have only

34:50one choice and that's not flexibility. So

34:52let's wait

34:53and see what's going on there. Linked tasks

34:56with prep conductor. So this was already

34:59released in

34:59the previous one but let's see easily

35:02automate the orchestration of multiple flow

35:05jobs ensuring that

35:06they happen in sequence. Building on the

35:08link task functionality launched in Tableau

35:11prep 2021.3,

35:12you can now link tasks to run on already

35:15scheduled flows that set back up flows to

35:18run in case of a

35:19failed run. Interesting. So link task was

35:23released previously. That was pretty

35:26straightforward but I

35:29think this capability with prep conductor

35:32starts to sort of take it to the next level

35:34. It's

35:34essentially an enhancement in my eyes I

35:36think rather than just necessarily a new

35:39feature. The

35:39two are sort of linked. Vis load times data

35:42source for admin insights. So this is going

35:46to essentially

35:46just give you the load times in admin

35:48insights. Again I believe this is Tableau

35:51online. Admin

35:52insights is a specific Tableau online data

35:55source. You can see that it says here and

35:57so that will be

35:58good to know so you can actually start to

35:59measure performance. That wasn't available.

36:01Tableau server

36:02users or server admins have this already so

36:05it's not new for them. Parameter

36:06enhancements in Tableau

36:08prep. So in a previous release they

36:10released a parameter enhancement. Now what

36:13they're doing in

36:14version 22.1 Tableau prep is adding even

36:16more places where you can use parameters in

36:18your flow

36:19as well as user enhancements. So now you

36:21can delete parameters with a single click,

36:24include parameters in SQL scripts and

36:26include parameter override values when

36:28running flows

36:29using the rest API. That's pretty handy. So

36:31you can programmatically trigger a flow

36:33using the

36:34rest API and pass in an argument that goes

36:35into the parameter. That's pretty cool.

36:38That's actually

36:38pretty dynamic because what you can do with

36:41that is you can build a flow that accepts

36:43variables

36:43but then does a bunch of calculations in a

36:46specific way just using an API, publishes

36:48the data to a

36:48specific place, emails or messages someone

36:51to say hey your data set is ready. They can

36:54go in,

36:55go into ask data and start asking questions

36:57from it. So that's pretty cool. It's

36:58probably not that

36:59smooth as these are sort of one of these

37:02perfect solutions as it were but

37:03nonetheless this is what

37:05you're getting. So this is nice. Right last

37:08few customer data platform accelerator in

37:10Tableau

37:11Online. Accelerate your time to data

37:13insights in this all-encompassing view of

37:15customers,

37:15engagement activities including email,

37:18mobile notifications and more.

37:19Interesting customer data platform

37:22accelerator. I don't know what that

37:24actually means and this

37:25description is not that clear so maybe one

37:28for the team I'll have to ask and find out

37:30more.

37:31Prep flow notifications in Slack. Again,

37:33Slack coming up again. Notifications going

37:38to

37:39Slack from Tableau related to flows. So

37:42that's pretty straightforward. Virtual

37:46connections for

37:46Salesforce. So this is good. Salesforce

37:49connections are really frustrating in Table

37:53au desktop and

37:53Tableau prep because they don't work

37:55exactly the same way. The virtual

37:57connection is going to

37:58standardize this a lot. I think this is one

38:00of the good areas where the virtual

38:01connection can

38:02actually help because by standardizing the

38:04connection what you can then do is you can

38:07sort

38:07of make have visibility of what people are

38:09doing, what people are using but you can

38:11stop this thing

38:11that's quite easy to do with Salesforce

38:13which is lots of people connecting and

38:15creating Salesforce

38:17views but they haven't done the due

38:19diligence to check how those views should

38:21come together

38:21and how necessarily those things work. So

38:24this will allow someone to create something

38:26,

38:26you know, verify it, give it a start and

38:28then lots of people can trust it and just

38:30get on with using

38:31it rather than worrying if they've talked

38:33to the admin to make sure they've got the

38:34right

38:35objects and fields from Salesforce to work

38:37together with Tableau.

38:38Right, last row. Improved OAuth login

38:41process. I think that sounds exactly like

38:44what it says.

38:45The process is better. I don't know much

38:48about OAuth. It's an authentication so what

38:51is OAuth?

38:51What does the term OAuth stand for? I've

38:53seen this term all the time but I've never

38:55actually...

38:56Open authorization. Okay, well yeah, but

38:58what does that actually mean? What I can

39:02say is when

39:02you're working with APIs, authorization is

39:05a really important concept and in essence

39:08the best

39:10way to think of it is let's say you want to

39:12log into Tableau but you don't want to use

39:14a manual

39:14Tableau login to type a username and

39:16password. What you have to do is Tableau

39:19has to, you know,

39:20trust another process to do that checking

39:21to make sure you're you. So if I'm going to

39:24build

39:24an application and I want the application

39:26to take charge of checking who you are, let

39:28's say the

39:29application is a mobile banking app, Table

39:31au in that instance is going to need to hand

39:34over the

39:34trust to the banking app to check that I am

39:37Tim. The banking app will need to do its

39:39own checks,

39:40logins and so on and so forth. It could

39:42even use an enterprise login. Now when that

39:45banking app

39:45says yes this is Tim, it passes back a

39:48token and an authorization and then the

39:50existing Tableau

39:51server uses that. In this case essentially

39:53what they're doing is improving that login

39:56experience,

39:56it's a little bit smoother but they're not

39:58really changing the infrastructure of how

40:00that works.

40:01Right, virtual connections integrated with

40:04Tableau catalog. Virtual connections and

40:08new

40:08content type, I think these are going to be

40:10available now in catalog so

40:13previously they weren't in catalog and

40:15catalog is supposed to be the metadata view

40:18of Tableau.

40:18So now that they're in there you can start

40:20to see what people are doing with them,

40:22where they've

40:22been used up and downstream and then of

40:25course the API will also assume have this

40:28capability.

40:29So that's pretty good. This will open up

40:31lineage capabilities that will shine a

40:34light on who or

40:35what is using specific virtual connections

40:37to speed up collaborations or reduce

40:39disruption

40:39when changes happen. So avoid breaking

40:42people's work by going in and looking at

40:45the virtual

40:45connections and seeing who's using them,

40:47that's pretty good. The last one for this

40:49is the embedding

40:50API version 3.1. Now this, what is this? So

40:54this is like a portal. What we're looking

40:58at here is

40:58like a portal and someone's built it.

41:00Essentially it's a custom Tableau portal.

41:02On the left hand

41:03side you've got a navigation that goes to

41:05different views and over here you have a

41:07window

41:07that loads up the dashboard and depending

41:09on what you click it loads the dashboard.

41:12Now what is

41:12interesting about this is that up here you

41:14have some things that aren't Tableau, so

41:16everything

41:17I'm highlighting in blue is not part of

41:19Tableau, it's been built into an

41:20application a bit like the

41:22banking app I was telling you about. And

41:25the embedding API is being improved to

41:27enable this

41:28to become a more seamless experience. If

41:31you want to check out some content on this

41:34go check out my

41:35friend Andre, he has a great YouTube

41:37channel. I'll put a link in the description

41:39so you can go check

41:40him out, put it at the top so you don't

41:41have to scroll too far for it. Go check him

41:43out because

41:43he covers all this stuff really really well

41:45and he's a developer so he actually

41:47understands

41:48the importance of why these things coming

41:50together makes a lot of sense. So that is

41:54it, that's all

41:55the features. We've gone through like a

41:57very sort of brief overview of what's

41:58coming soon. Starting

42:00probably next week or this week, I'll start

42:02making videos on the features that are

42:04available already

42:05in Tableau Online. But other than that,

42:07thank you for watching and if you've made

42:09it all the way to

42:09the very end let me know what you thought

42:12of this video and the features coming in

42:142022.1.

42:15And yeah, I'll catch you in the next one.

42:19Take it easy.

42:20you

42:24you

42:25you

42:25you

42:26[ Silence ]